ALBERT

Description: Given a database table with records that can be ranked, an interesting problem is to identify selection conditions for the table, which are qualified by an input record and render its ranking as high as possible among the qualifying tuples. In this paper, we study this standing maximization problem, which finds application in object promotion and characterization. After showing the hardness of the problem, we propose greedy methods, which are experimentally shown to achieve high accuracy compared to exhaustive enumeration, while scaling very well to the problem input size. Our contributions include a linear-time algorithm for determining the optimal selection range for an ordinal attribute and techniques for choosing and prioritizing the most promising selection predicates to apply. Experiments on real datasets confirm the effectiveness and efficiency of our techniques.

Description: Some fairly recent research has focused on providing XACML-based solutions for dynamic privacy policy management. In this regard, a number of works have provided enhancements to the performance of XACML policy enforcement point (PEP) component, but very few have focused on enhancing the accuracy of that component. This paper improves the accuracy of an XACML PEP by filling some gaps in the existing works. In particular, dynamically incorporating user access context into the privacy policy decision, and its enforcement. We provide an XACML-based implementation of a dynamic privacy policy management framework and an evaluation of the applicability of our system in comparison to some of the existing approaches.

Description: This paper first introduces pattern aided regression (PXR) models, a new type of regression models designed to represent accurate and interpretable prediction models. This was motivated by two observations: (1) Regression modeling applications often involve complex diverse predictor-response relationships , which occur when the optimal regression models (of given regression model type) fitting two or more distinct logical groups of data are highly different. (2) State-of-the-art regression methods are often unable to adequately model such relationships. This paper defines PXR models using several patterns and local regression models, which respectively serve as logical and behavioral characterizations of distinct predictor-response relationships. The paper also introduces a contrast pattern aided regression (CPXR) method, to build accurate PXR models. In experiments, the PXR models built by CPXR are very accurate in general, often outperforming state-of-the-art regression methods by big margins. Usually using (a) around seven simple patterns and (b) linear local regression models, those PXR models are easy to interpret; in fact, their complexity is just a bit higher than that of (piecewise) linear regression models and is significantly lower than that of traditional ensemble based regression models. CPXR is especially effective for high-dimensional data. The paper also discusses how to use CPXR methodology for analyzing prediction models and correcting their prediction errors.

Description: We analyze models for predicting the probability of a strikeout for a batter/pitcher matchup in baseball using player descriptors that can be estimated accurately from small samples. We start with the log5 model which has been used extensively for describing matchups in sports. Log5 is a special case of a logit model and we use constrained logistic regression over nearly one million matchup observations to assess the use of the log5 explanatory variables for this application. We also show that a batter/pitcher ground ball rate interaction variable is significant for the prediction of strikeout probability and we provide physical justification for the inclusion of this variable in the model. We quantify the differences among the models and show that batters control the majority of the variance in predicted strikeout rate.

Description: Over the past decade or so, several research groups have addressed the problem of multi-label classification where each example can belong to more than one class at the same time. A common approach, called  Binary Relevance (BR) , addresses this problem by inducing a separate classifier for each class. Research has shown that this framework can be improved if mutual class dependence is exploited: an example that belongs to class $X$ is likely to belong also to class $Y$ ; conversely, belonging to $X$ can make an example less likely to belong to $Z$ . Several works sought to model this information by using the vector of class labels as additional example attributes. To fill the unknown values of these attributes during prediction, existing methods resort to using outputs of other classifiers, and this makes them prone to errors. This is where our paper wants to contribute. We identified two potential ways to prune unnecessary dependencies and to reduce error-propagation in our new classifier-stacking technique, which is named PruDent . Experimental results indicate that the classification performance of PruDent compares favorably with that of other state-of-the-art approaches over a broad range of testbeds. Mor- over, its computational costs grow only linearly in the number of classes.

Description: This work deals with the problem of producing a fast and accurate data classification, learning it from a possibly small set of records that are already classified. The proposed approach is based on the framework of the so-called Logical Analysis of Data (LAD), but enriched with information obtained from statistical considerations on the data. A number of discrete optimization problems are solved in the different steps of the procedure, but their computational demand can be controlled. The accuracy of the proposed approach is compared to that of the standard LAD algorithm, of support vector machines and of label propagation algorithm on publicly available datasets of the UCI repository. Encouraging results are obtained and discussed.

Description: A new graph based constrained semi-supervised learning (G-CSSL) framework is proposed. Pairwise constraints (PC) are used to specify the types (intra- or inter-class) of points with labels. Since the number of labeled data is typically small in SSL setting, the core idea of this framework is to create and enrich the PC sets using the propagated soft labels from both labeled and unlabeled data by special label propagation (SLP), and hence obtaining more supervised information for delivering enhanced performance. We also propose a Two-stage Sparse Coding, termed TSC, for achieving adaptive neighborhood for SLP. The first stage aims at correcting the possible corruptions in data and training an informative dictionary, and the second stage focuses on sparse coding. To deliver enhanced inter-class separation and intra-class compactness, we also present a mixed soft-similarity measure to evaluate the similarity/dissimilarity of constrained pairs using the sparse codes and outputted probabilistic values by SLP. Simulations on the synthetic and real datasets demonstrated the validity of our algorithms for data representation and image recognition, compared with other related state-of-the-art graph based semi-supervised techniques.

Description: In large databases, the amount and the complexity of the data calls for data summarization techniques. Such summaries are used to assist fast approximate query answering or query optimization. Histograms are a prominent class of model-free data summaries and are widely used in database systems. So-called self-tuning histograms look at query-execution results to refine themselves. An assumption with such histograms, which has not been questioned so far, is that they can learn the dataset from scratch, that is—starting with an empty bucket configuration. We show that this is not the case. Self-tuning methods are very sensitive to the initial configuration. Three major problems stem from this. Traditional self-tuning is unable to learn projections of multi-dimensional data, is sensitive to the order of queries, and reaches only local optima with high estimation errors. We show how to improve a self-tuning method significantly by starting with a carefully chosen initial configuration. We propose initialization by dense subspace clusters in projections of the data, which improves both accuracy and robustness of self-tuning. Our experiments on different datasets show that the error rate is typically halved compared to the uninitialized version.

Description: Recently, two ideas have been explored that lead to more accurate algorithms for time-series classification (TSC). First, it has been shown that the simplest way to gain improvement on TSC problems is to transform into an alternative data space where discriminatory features are more easily detected. Second, it was demonstrated that with a single data representation, improved accuracy can be achieved through simple ensemble schemes. We combine these two principles to test the hypothesis that forming a collective of ensembles of classifiers on different data transformations improves the accuracy of time-series classification. The collective contains classifiers constructed in the time, frequency, change, and shapelet transformation domains. For the time domain, we use a set of elastic distance measures. For the other domains, we use a range of standard classifiers. Through extensive experimentation on 72 datasets, including all of the 46 UCR datasets, we demonstrate that the simple collective formed by including all classifiers in one ensemble is significantly more accurate than any of its components and any other previously published TSC algorithm. We investigate alternative hierarchical collective structures and demonstrate the utility of the approach on a new problem involving classifying Caenorhabditis elegans mutant types.

Description: In real-world graphs such as social networks, Semantic Web and biological networks, each vertex usually contains rich information, which can be modeled by a set of tokens or elements. In this paper, we study a subgraph matching with set similarity (SMS $^2$ ) query over a large graph database, which retrieves subgraphs that are structurally isomorphic to the query graph, and meanwhile satisfy the condition of vertex pair matching with the (dynamic) weighted set similarity. To efficiently process the SMS $^2$ query, this paper designs a novel lattice-based index for data graph, and lightweight signatures for both query vertices and data vertices. Based on the index and signatures, we propose an efficient two-phase pruning strategy including set similarity pruning and structure-based pruning, which exploits the unique features of both (dynamic) weighted set similarity and graph topology. We also propose an efficient dominating-set-based subgraph matching algorithm guided by a dominating set selection algorithm to achieve better query performance. Extensive experiments on both real and synthetic datasets demonstrate that our method outperforms state-of-the-art methods by an order of magnitude.

Description: Data imputation aims at filling in missing attribute values in databases. Most existing imputation methods to string attribute values are inferring-based approaches, which usually fail to reach a high imputation recall by just inferring missing values from the complete part of the data set. Recently, some retrieving-based methods are proposed to retrieve missing values from external resources such as the World Wide Web, which tend to reach a much higher imputation recall, but inevitably bring a large overhead by issuing a large number of search queries. In this paper, we investigate the interaction between the inferring-based methods and the retrieving-based methods. We show that retrieving a small number of selected missing values can greatly improve the imputation recall of the inferring-based methods. With this intuition, we propose an inTeractive Retrieving-Inferring data imPutation approach (TRIP), which performs retrieving and inferring alternately in filling in missing attribute values in a data set. To ensure the high recall at the minimum cost, TRIP faces a challenge of selecting the least number of missing values for retrieving to maximize the number of inferable values. Our proposed solution is able to identify an optimal retrieving-inferring scheduling scheme in deterministic data imputation, and the optimality of the generated scheme is theoretically analyzed with proofs. We also analyze with an example that the optimal scheme is not feasible to be achieved in $tau$ -constrained stochastic data imputation ( $tau$ -SDI), but still, our proposed solution identifies an expected-optimal scheme in $tau$ -SDI. Extensive experiments on four data collections show that TRIP retrieves on average 20 percent missing values and achieves the same high recall that was reached by the retrieving-based approach.

Description: Visual classification has attracted considerable research interests in the past decades. In this paper, a novel $ell _1$ -hypergraph model for visual classification is proposed. Hypergraph learning, as a natural extension of graph model, has been widely used in many machine learning tasks. In previous work, hypergraph is usually constructed by attribute-based or neighborhood-based methods. That is, a hyperedge is generated by connecting a set of samples sharing a same feature attribute or in a neighborhood. However, these methods are unable to explore feature space globally or sensitive to noises. To address these problems, we propose a novel hypergraph construction approach that leverages sparse representation to generate hyperedges and learns the relationship among hyperedges and their vertices. First, for each sample, a hyperedge is generated by regarding it as the centroid and linking it as well as its nearest neighbors. Then, the sparse representation method is applied to represent the centroid vertex by other vertices within the same hyperedge. The vertices with zero coefficients are removed from the hyperedge. Finally, the representation coefficients are used to define the incidence relation between the hyperedge and the vertices. In our approach, we also optimize the hyperedge weights to modulate the effects of different hyperedges. We leverage the prior knowledge on the hyperedges so that the hyperedges sharing more vertices can have closer weights, where a graph Laplacian is used to regularize the optimization of the weights. Our approach is named $ell _1$ -hypergraph since the $ell _1$ sparse representation is employed in the hypergraph construction process. The method is evaluated on various visual classification tasks, and it demonstrates promising performance.

Description: We consider the problem of adaptively routing a fleet of cooperative vehicles within a road network in the presence of uncertain and dynamic congestion conditions. To tackle this problem, we first propose a Gaussian process dynamic congestion model that can effectively characterize both the dynamics and the uncertainty of congestion conditions. Our model is efficient and thus facilitates real-time adaptive routing in the face of uncertainty. Using this congestion model, we develop efficient algorithms for non-myopic adaptive routing to minimize the collective travel time of all vehicles in the system. A key property of our approach is the ability to efficiently reason about the long-term value of exploration, which enables collectively balancing the exploration/exploitation trade-off for entire fleets of vehicles. Our approach is validated by traffic data from two large Asian cities. Our congestion model is shown to be effective in modeling dynamic congestion conditions. Our routing algorithms also generate significantly faster routes compared to standard baselines, and achieve near-optimal performance compared to an omniscient routing algorithm. We also present the results from a preliminary field study, which showcases the efficacy of our approach.

Description: Betweenness centrality is a classic measure that quantifies the importance of a graph element (vertex or edge) according to the fraction of shortest paths passing through it. This measure is notoriously expensive to compute, and the best known algorithm runs in $mathcal {O}(nm)$ time. The problems of efficiency and scalability are exacerbated in a dynamic setting, where the input is an evolving graph seen edge by edge, and the goal is to keep the betweenness centrality up to date. In this paper, we propose the first truly scalable algorithm for online computation of betweenness centrality of both vertices and edges in an evolving graph where new edges are added and existing edges are removed. Our algorithm is carefully engineered with out-of-core techniques and tailored for modern parallel stream processing engines that run on clusters of shared-nothing commodity hardware. Hence, it is amenable to real-world deployment. We experiment on graphs that are two orders of magnitude larger than previous studies. Our method is able to keep the betweenness centrality measures up-to-date online, i.e., the time to update the measures is smaller than the inter-arrival time between two consecutive updates.

Description: Phase change memory (PCM) is non-volatile memory that is byte-addressable. It is two to four times denser than DRAM, orders of magnitude better than NAND Flash memory in read latency, and 10 times better than NAND Flash memory in write endurance. However, it still limits the number of write operations to at most $10^6$ times per PCM cell. To extend its lifetime, it is necessary to evenly distribute write operations over all the memory cells. Up to now, the $mathrm{B^{+}}$ -Tree index structure has been used to quickly locate a search key in a relational database management system (RDBMS). All the record keys in each node are sorted and packed upon insertion in, and deletion from, the $mathrm{B^{+}}$ -Tree. In addition, a counter keeps track of the number of valid keys in the $mathrm{B^{+}}$ -Tree. Consequently, a $mathrm{B^{+}}$ -Tree algorithm results in a large number of write operations, which deteriorates the endurance of PCM. This restricts the usage of PCM on a database server and deteriorates performance of database servers. In this paper, we propose a novel PCM-aware $math- m{B^{+}}$ -Tree index structure, called $mathrm{PB^{+}}$ -Tree, to provide wear-leveling in PCM. According to our experiment results, $mathrm{PB^{+}}$ -Tree is much faster than the existing $mathrm{B^{+}}$ -Tree algorithms for PCM and NAND Flash memory with versatile workloads. More importantly, our scheme also greatly reduces the number of write operations compared to other $mathrm{B^{+}}$ -Tree algorithms. All of these results suggest that $mathrm{PB^{+}}$ -Tree is the $mathrm{B^{+}}$ -Tree algorithm best fitted to PCM.

Description: The $k$ nearest neighbor ( $k$ NN) search on road networks is an important function in web mapping services. These services are now dealing with rapidly arriving queries, that are issued by a massive amount of users. While overlay graph-based indices can answer shortest path queries efficiently, there have been no studies on utilizing such indices to answer $k$ NN queries efficiently. In this paper, we fill this research gap and present two efficient $k$ NN search solutions on overlay graph-based indices. Experimental results show that our solutions offer very low query latency (0.1 ms) and require only small index sizes, even for 10-million-node networks.

Description: Measuring semantic similarity between two terms is essential for a variety of text analytics and understanding applications. Currently, there are two main approaches for this task, namely the knowledge based and the corpus based approaches. However, existing approaches are more suitable for semantic similarity between words rather than the more general multi-word expressions (MWEs), and they do not scale very well. Contrary to these existing techniques, we propose an efficient and effective approach for semantic similarity using a large scale semantic network. This semantic network is automatically acquired from billions of web documents. It consists of millions of concepts, which explicitly model the context of semantic relationships. In this paper, we first show how to map two terms into the concept space, and compare their similarity there. Then, we introduce a clustering approach to orthogonalize the concept space in order to improve the accuracy of the similarity measure. Finally, we conduct extensive studies to demonstrate that our approach can accurately compute the semantic similarity between terms of MWEs and with ambiguity, and significantly outperforms 12 competing methods under Pearson Correlation Coefficient. Meanwhile, our approach is much more efficient than all competing algorithms, and can be used to compute semantic similarity in a large scale.

Description: Given a spatio-temporal network, a source, a destination, and a desired departure time interval, the All-departure-time Lagrangian Shortest Paths (ALSP) problem determines a set which includes the shortest path for every departure time in the given interval. ALSP is important for critical societal applications such as eco-routing. However, ALSP is computationally challenging due to the non-stationary ranking of the candidate paths across distinct departure-times. Current related work for reducing the redundant work, across consecutive departure-times sharing a common solution, exploits only partial information e.g., the earliest feasible arrival time of a path. In contrast, our approach uses all available information, e.g., the entire time series of arrival times for all departure-times. This allows elimination of all knowable redundant computation based on complete information available at hand. We operationalize this idea through the concept of critical-time-points (CTP), i.e., departure-times before which ranking among candidate paths cannot change. In our preliminary work, we proposed a CTP based forward search strategy. In this paper, we propose a CTP based temporal bi-directional search for the ALSP problem via a novel impromptu rendezvous termination condition. Theoretical and experimental analysis show that the proposed approach outperforms the related work approaches particularly when there are few critical-time-points.

Description: Computing connected components is a core operation on graph data. Since billion-scale graphs cannot be resident in memory of a single server, several approaches based on distributed machines have recently been proposed. The representative methods are $mathsf{Hashhbox{-}Tohbox{-}Min}$ and $mathsf{PowerGraph}$ . $mathsf{Hashhbox{-}Tohbox{-}Min}$ is the state-of-the art disk-based distributed method which minimizes the number of MapReduce rounds. $mathsf{PowerGraph}$ is the-state-of-the-art in-memory distributed system, which is typically faster than the disk-based distributed one, however, requires a lot of machines for handling billion-scale graphs. In this paper, we propose an I/O efficient parallel algorithm for billion-scale graphs in a single PC. We first propose the Disk-based Sequential access-oriented Parallel processing  (DSP) model that exploits sequential disk access in terms of disk I/Os and parallel processing in terms of computation. We then propose an ultra-fast disk-based parallel algorithm for computing connected components, $mathsf{DSPhbox{-}CC}$ , which largely improves the performance through sequential disk scan and page-level cache-conscious parallel processing . Extensive experimental results show that $mathsf{DSPhbox{-}CC}$ 1) computes connected components in billion-scale graphs using the limited memory size whereas in-memory algorithms can only support medium-sized graphs with the same memory size, and 2) significantly outperforms all distributed competitors as well as a representative disk-based parallel method.

Description: Answering why-not questions in databases is promised to have wide application prospect in many areas and thereby, has attracted recent attention in the database research community. This paper addresses the problem of answering these so-called why-not questions in similar graph matching for graph databases. Given a set of answer graphs of an initial query graph $q$ and a set of missing ( why-not ) graphs, we aim to modify $q$ into a new query graph $q^*$ such that the missing graphs are included in the new answer set of $q^*$ . We present an approximate solution to address the above as the optimal solution is NP-hard to compute. In our approach, we first compute the bounded search space and the distance to be minimized for $q^*$ . Then, we present a two-phase algorithm to find the new query $q^*$ . In the first phase, we generate a set of candidate edges to be added/deleted into/from the initial query $q$ within the bounded search space and in the second phase, we select a subset of candidate edges generated in the first phase to minimize the distance for $q^*$ . We also demonstrate the effectiveness and efficiency of our approach by conducting extensive experiments on two real datasets.

Description: How has the interdisciplinary data mining field been practiced in Network and Systems Management (NSM)? In Science and Technology, there is a wide use of data mining in areas like bioinformatics, genetics, Web, and, more recently, astroinformatics. However, the application in NSM has been limited and inconsiderable. In this article, we provide an account of how data mining has been applied in managing networks and systems for the past four decades, presumably since its birth. We look into the field’s applications in the key NSM activities—discovery, monitoring, analysis, reporting, and domain knowledge acquisition. In the end, we discuss our perspective on the issues that are considered critical for the effective application of data mining in the modern systems which are characterized by heterogeneity and high dynamism.

Description: With the rapid development of location-aware mobile devices, ubiquitous Internet access and social computing technologies, lots of users’ personal information, such as location data and social data, has been readily accessible from various mobile platforms and online social networks. The convergence of these two types of data, known as geo-social data , has enabled collaborative spatial computing that explicitly combines both location and social factors to answer useful geo-social queries for either business or social good. In this paper, we study a new type of Geo-Social K-Cover Group (GSKCG) queries that, given a set of query points and a social network, retrieves a minimum user group in which each user is socially related to at least $k$ other users and the users’ associated regions (e.g., familiar regions or service regions) can jointly cover all the query points. Albeit its practical usefulness, the GSKCG query problem is NP-complete. We consequently explore a set of effective pruning strategies to derive an efficient algorithm for finding the optimal solution. Moreover, we design a novel index structure tailored to our problem to further accelerate query processing. Extensive experiments demonstrate that our algorithm achieves desirable performance on real-life datasets.

Description: High utility sequential pattern mining has been considered as an important research problem and a number of relevant algorithms have been proposed for this topic. The main challenge of high utility sequential pattern mining is that, the search space is large and the efficiency of the solutions is directly affected by the degree at which they can eliminate the candidate patterns. Therefore, the efficiency of any high utility sequential pattern mining solution depends on its ability to reduce this big search space, and as a result, lower the computational complexity of calculating the utilities of the candidate patterns. In this paper, we propose efficient data structures and pruning technique which is based on Cumulated Rest of Match (CRoM) based upper bound. CRoM, by defining a tighter upper bound on the utility of the candidates, allows more conservative pruning before candidate pattern generation in comparison to the existing techniques. In addition, we have developed an efficient algorithm, High Utility Sequential Pattern Extraction (HuspExt), which calculates the utilities of the child patterns based on that of the parents’. Substantial experiments on both synthetic and real datasets from different domains show that, the proposed solution efficiently discovers high utility sequential patterns from large scale datasets with different data characteristics, under low utility thresholds.

Description: Ordinal classification with a monotonicity constraint is a kind of classification tasks, in which the objects with better attribute values should not be assigned to a worse decision class. Several learning algorithms have been proposed to handle this kind of tasks in recent years. The rank entropy-based monotonic decision tree is very representative thanks to its better robustness and generalization. Ensemble learning is an effective strategy to significantly improve the generalization ability of machine learning systems. The objective of this work is to develop a method of fusing monotonic decision trees. In order to achieve this goal, we take two factors into account: attribute reduction and fusing principle. Through introducing variable dominance rough sets, we firstly propose an attribute reduction approach with rank-preservation for learning base classifiers, which can effectively avoid overfitting and improve classification performance. Then, we establish a fusing principe based on maximal probability through combining the base classifiers, which is used to further improve generalization ability of the learning system. The experimental analysis shows that the proposed fusing method can significantly improve classification performance of the learning system constructed by monotonic decision trees.

Description: Influence maximization, defined as finding a small subset of nodes that maximizes spread of influence in social networks, is NP-hard under both Independent Cascade (IC) and Linear Threshold (LT) models, where many greedy-based algorithms have been proposed with the best approximation guarantee. However, existing greedy-based algorithms are inefficient on large networks, as it demands heavy Monte-Carlo simulations of the spread functions for each node at the initial step [7] . In this paper, we establish new upper bounds to significantly reduce the number of Monte-Carlo simulations in greedy-based algorithms, especially at the initial step. We theoretically prove that the bound is tight and convergent when the summation of weights towards (or from) each node is less than 1. Based on the bound, we propose a new Upper Bound based Lazy Forward algorithm ( UBLF in short) for discovering the top-k influential nodes in social networks. We test and compare UBLF with prior greedy algorithms, especially CELF [30] . Experimental results show that UBLF reduces more than 95 percent Monte-Carlo simulations of CELF and achieves about $2hbox{-}10$ times speedup when the seed set is small.

Description: Feature selection has been an important research topic in data mining, because the real data sets often have high-dimensional features, such as the bioinformatics and text mining applications. Many existing filter feature selection methods rank features by optimizing certain feature ranking criterions, such that correlated features often have similar rankings. These correlated features are redundant and don’t provide large mutual information to help data mining. Thus, when we select a limited number of features, we hope to select the top non-redundant features such that the useful mutual information can be maximized. In previous research, Ding et al. recognized this important issue and proposed the minimum Redundancy Maximum Relevance Feature Selection (mRMR) model to minimize the redundancy between sequentially selected features. However, this method used the greedy search, thus the global feature redundancy wasn’t considered and the results are not optimal. In this paper, we propose a new feature selection framework to globally minimize the feature redundancy with maximizing the given feature ranking scores, which can come from any supervised or unsupervised methods. Our new model has no parameter so that it is especially suitable for practical data mining application. Experimental results on benchmark data sets show that the proposed method consistently improves the feature selection results compared to the original methods. Meanwhile, we introduce a new unsupervised global and local discriminative feature selection method which can be unified with the global feature redundancy minimization framework and shows superior performance.

Description: Multimedia information retrieval usually involves two key modules including effective feature representation and ranking model construction. Most existing approaches are incapable of well modeling the inherent correlations and interactions between them, resulting in the loss of the latent consensus structure information. To alleviate this problem, we propose a learning to rank approach that simultaneously obtains a set of deep linear features and constructs structure-aware ranking models in a joint learning framework. Specifically, the deep linear feature learning corresponds to a series of matrix factorization tasks in a hierarchical manner, while the learning-to-rank part concentrates on building a ranking model that effectively encodes the intrinsic ranking information by structural SVM learning. Through a joint learning mechanism, the two parts are mutually reinforced in our approach, and meanwhile their underlying interaction relationships are implicitly reflected by solving an alternating optimization problem. Due to the intrinsic correlations among different queries (i.e., similar queries for similar ranking lists), we further formulate the learning-to-rank problem as a multi-task problem, which is associated with a set of mutually related query-specific learning-to-rank subproblems. For computational efficiency and scalability, we design a MapReduce-based parallelization approach to speed up the learning processes. Experimental results demonstrate the efficiency, effectiveness, and scalability of the proposed approach in multimedia information retrieval.

Description: In a peer-to-peer system, a node should estimate reputation of other peers not only on the basis of its own interaction, but also on the basis of expression of other nodes. Reputation aggregation mechanism implements strategy for achieving this. Reputation aggregation in peer to peer networks is generally a very time and resource consuming process. Moreover, most of the methods consider that a node will have the same reputation after aggregation with all the nodes in the network, which is not true. This paper proposes a reputation aggregation algorithm that uses a variant of gossip algorithm called differential gossip. In this paper, estimate of reputation is considered to be having two parts, one common component which is same with every node, and the other one is the information received from immediate neighbours based on the neighbours’ direct interaction with the node. The differential gossip is fast and requires a lesser amount of resources. This mechanism allows computation of independent reputation value by every node, of every other node in the network. The differential gossip trust has been investigated for a power law network formed using preferential attachment (PA) Model. The reputation computed using differential gossip trust shows good amount of immunity to the collusion. We have verified the performance of the algorithm on the power law networks with sizes ranging from 100 nodes to 50,000 nodes.

Description: In this paper, the limitation that is prominent in most existing works of change-point detection methods is addressed by proposing a nonparametric, computationally efficient method. The limitation is that most works assume that each data point observed at each time step is a single multi-dimensional vector. However, there are many situations where this does not hold. Therefore, a setting where each observation is a collection of random variables, which we call a bag of data, is considered. After estimating the underlying distribution behind each bag of data and embedding those distributions in a metric space, the change-point score is derived by evaluating how the sequence of distributions is fluctuating in the metric space using a distance-based information estimator. Also, a procedure that adaptively determines when to raise alerts is incorporated by calculating the confidence interval of the change-point score at each time step. This avoids raising false alarms in highly noisy situations and enables detecting changes of various magnitudes. A number of experimental studies and numerical examples are provided to demonstrate the generality and the effectiveness of our approach with both synthetic and real datasets.

Description: In many applications, top- k query is an important operation to return a set of interesting points in a potentially huge data space. It is analyzed in this paper that the existing algorithms cannot process top- k query on massive data efficiently. This paper proposes a novel table-scan-based T2S algorithm to efficiently compute top- k results on massive data. T2S first constructs the presorted table, whose tuples are arranged in the order of the round-robin retrieval on the sorted lists. T2S maintains only fixed number of tuples to compute results. The early termination checking for T2S is presented in this paper, along with the analysis of scan depth. The selective retrieval is devised to skip the tuples in the presorted table which are not top- k results. The theoretical analysis proves that selective retrieval can reduce the number of the retrieved tuples significantly. The construction and incremental-update/batch-processing methods for the used structures are proposed in this paper. The extensive experimental results, conducted on synthetic and real-life data sets, show that T2S has a significant advantage over the existing algorithms.

Description: We witness an unprecedented proliferation of knowledge graphs that record millions of entities and their relationships. While knowledge graphs are structure-flexible and content-rich, they are difficult to use. The challenge lies in the gap between their overwhelming complexity and the limited database knowledge of non-professional users. If writing structured queries over “simple” tables is difficult, complex graphs are only harder to query. As an initial step toward improving the usability of knowledge graphs, we propose to query such data by example entity tuples, without requiring users to form complex graph queries. Our system, Graph Query By Example ( $mathsf {GQBE}$ ), automatically discovers a weighted hidden maximum query graph based on input query tuples, to capture a user’s query intent. It then efficiently finds and ranks the top approximate matching answer graphs and answer tuples. We conducted experiments and user studies on the large Freebase and DBpedia datasets and observed appealing accuracy and efficiency. Our system provides a complementary approach to the existing keyword-based methods, facilitating user-friendly graph querying. To the best of our knowledge, there was no such proposal in the past in the context of graphs.

Description: This paper explores combinatorial optimization for problems of max-weight graph matching on multi-partite graphs, which arise in integrating multiple data sources. In the most common two-source case, it is often desirable for the final matching to be one-to-one; the database and statistical record linkage communities accomplish this by weighted bipartite graph matching on similarity scores. Such matchings are intuitively appealing: they leverage a natural global property of many real-world entity stores—that of being nearly deduped—and are known to provide significant improvements to precision and recall. Unfortunately, unlike the bipartite case, exact max-weight matching on multi-partite graphs is known to be NP-hard. Our two-fold algorithmic contributions approximate multi-partite max-weight matching: our first algorithm borrows optimization techniques common to Bayesian probabilistic inference; our second is a greedy approximation algorithm. In addition to a theoretical guarantee on the latter, we present comparisons on a real-world entity resolution problem from Bing significantly larger than typically found in the literature, on publication data, and on a series of synthetic problems. Our results quantify significant improvements due to exploiting multiple sources, which are made possible by global one-to-one constraints linking otherwise independent matching sub-problems. We also discover that our algorithms are complementary: one being much more robust under noise, and the other being simple to implement and very fast to run.

Description: Internet users can be classified as two types: (a) active users —actively contribute to blogs, publish opinions, write comments in Youtube, tweet messages, etc. and (b) passive consumers who only consume Internet information without contributing to it. While the majority of current social media research deals with active user analysis, there is very little work in understanding the dynamics of passive consumers and their influence. Our global scale Internet measurement of user access patterns of a diverse set of Internet media services indicates conclusively that majority of consumers are passive. In this paper, we develop a spatio-temporal mathematical model and the corresponding stochastic analysis to understand the passive consumer dynamics. Both discrete and continuous time analysis are presented. We also show how the analysis can be used to identify spatial points of influence, i.e., spatial locations that have maximum expertise or influence on a topic. The analysis takes into account, the initial level of consumption at each spatial location and the influence of different passive consumers at different geographic locations on each other. The effect of information noise is taken into account to derive fundamental limits of passive information consumption. Theoretical results are verified using real Internet measurement data. The large scale data have been made available for other researchers.

Description: The objective of the paper is a contribution to data mining within the framework of the observational calculus, through introducing ǵeneralized quantifiers related to copulas. Fitting copulas to multidimensional data is an increasingly important method for analyzing dependencies, and the proposed quantifiers of observational calculus assess the results of estimating the structure of joint distributions of continuous variables by means of hierarchical Archimedean copulas. To this end, the existing theory of hierarchical Archimedean copulas has been slightly extended in the paper: It has been proven that sufficient conditions for the function defining a hierarchical Archimedean copula to be indeed a copula, which have so far been rigorously established only for the special case of fully nested Archimedean copulas, hold in general. These conditions allow us to define three new generalized quantifiers, which are then thoroughly validated on four benchmark data sets and one data set from a real-world application. The paper concludes by comparing the proposed quantifiers to a more traditional approach—maximum weight spanning trees.

Description: In this work, we develop an approach for the safe distribution and parallel execution of data-centric workflows over the publish/subscribe abstraction. In essence, we design a unique representation of data-centric workflows, specifically designed to exploit the loosely coupled and distributed nature of publish/subscribe systems. Furthermore, we argue for the practicality and expressiveness of our approach by mapping a standard and industry-strength data-centric workflow model, namely, IBM Business Artifacts with Guard-Stage-Milestone (GSM), into the publish/subscribe abstraction. In short, the contributions of this work are three-fold: (1) mapping of data-centric workflows into publish/subscribe to achieve distributed and parallel execution; (2) detailed theoretical analysis of the mapping; and (3) formulation of the complexity of the optimal workflow distribution over the publish/subscribe abstraction as an NP-hard problem.

Description: Domain transfer learning generalizes a learning model across training data and testing data with different distributions. A general principle to tackle this problem is reducing the distribution difference between training data and testing data such that the generalization error can be bounded. Current methods typically model the sample distributions in input feature space, which depends on nonlinear feature mapping to embody the distribution discrepancy. However, this nonlinear feature space may not be optimal for the kernel-based learning machines. To this end, we propose a transfer kernel learning (TKL) approach to learn a domain-invariant kernel by directly matching source and target distributions in the reproducing kernel Hilbert space (RKHS). Specifically, we design a family of spectral kernels by extrapolating target eigensystem on source samples with Mercer’s theorem. The spectral kernel minimizing the approximation error to the ground truth kernel is selected to construct domain-invariant kernel machines. Comprehensive experimental evidence on a large number of text categorization, image classification, and video event recognition datasets verifies the effectiveness and efficiency of the proposed TKL approach over several state-of-the-art methods.

Description: We present an evolutionary multi-branch tree clustering method to model hierarchical topics and their evolutionary patterns over time. The method builds evolutionary trees in a Bayesian online filtering framework. The tree construction is formulated as an online posterior estimation problem, which well balances both the fitness of the current tree and the smoothness between trees. The state-of-the-art multi-branch clustering method, Bayesian rose trees, is employed to generate a topic tree with a high fitness value. A constraint model is also introduced to preserve the smoothness between trees. A set of comprehensive experiments on real world news data demonstrates that the proposed method better incorporates historical tree information and is more efficient and effective than the traditional evolutionary hierarchical clustering algorithm. In contrast to our previous method [31] , we implement two additional baseline algorithms to compare them with our algorithm. We also evaluate the performance of the clustering algorithm based on multiple constraint trees. Furthermore, two case studies are conducted to demonstrate the effectiveness and usefulness of our algorithm in helping users understand the major hierarchical topic evolutionary patterns in text data.

Description: The discovery of regions of interest in large cities is an important challenge. We propose and investigate a novel query called the path nearby cluster (PNC) query that finds regions of potential interest (e.g., sightseeing places and commercial districts) with respect to a user-specified travel route. Given a set of spatial objects $O$ (e.g., POIs, geo-tagged photos, or geo-tagged tweets) and a query route $q$ , if a cluster $c$ has high spatial-object density and is spatially close to $q$ , it is returned by the query (a cluster is a circular region defined by a center and a radius). This query aims to bring important benefits to users in popular applications such as trip planning and location recommendation. Efficient computation of the PNC query faces two challenges: how to prune the search space during query processing, and how to identify clusters with high density effectively. To address these challenges, a novel collective search algorithm is developed. Conceptually, the search process is conducted in the spatial and density domains concurrently. In the spatial domain, network expansion is adopted, and a set of vertices are selected from the query route as expansion centers. In the density domain, clusters are sorted according to their density distributions and they are scan- ed from the maximum to the minimum. A pair of upper and lower bounds are defined to prune the search space in the two domains globally. The performance of the PNC query is studied in extensive experiments based on real and synthetic spatial data.

Description: Given learning samples from a raster data set, spatial decision tree learning aims to find a decision tree classifier that minimizes classification errors as well as salt-and-pepper noise. The problem has important societal applications such as land cover classification for natural resource management. However, the problem is challenging due to the fact that learning samples show spatial autocorrelation in class labels, instead of being independently identically distributed. Related work relies on local tests (i.e., testing feature information of a location) and cannot adequately model the spatial autocorrelation effect, resulting in salt-and-pepper noise. In contrast, we recently proposed a focal-test-based spatial decision tree (FTSDT), in which the tree traversal direction of a sample is based on both local and focal (neighborhood) information. Preliminary results showed that FTSDT reduces classification errors and salt-and-pepper noise. This paper extends our recent work by introducing a new focal test approach with adaptive neighborhoods that avoids over-smoothing in wedge-shaped areas. We also conduct computational refinement on the FTSDT training algorithm by reusing focal values across candidate thresholds. Theoretical analysis shows that the refined training algorithm is correct and more scalable. Experiment results on real world data sets show that new FTSDT with adaptive neighborhoods improves classification accuracy, and that our computational refinement significantly reduces training time.

Description: As a newly emerging network model, heterogeneous information networks (HINs) have received growing attention. Many data mining tasks have been explored in HINs, including clustering, classification, and similarity search. Similarity join is a fundamental operation required for many problems. It is attracting attention from various applications on network data, such as friend recommendation, link prediction, and online advertising. Although similarity join has been well studied in homogeneous networks, it has not yet been studied in heterogeneous networks. Especially, none of the existing research on similarity join takes different semantic meanings behind paths into consideration and almost all completely ignore the heterogeneity and diversity of the HINs. In this paper, we propose a path-based similarity join (PS-join) method to return the top $k$ similar pairs of objects based on any user specified join path in a heterogeneous information network. We study how to prune expensive similarity computation by introducing bucket pruning based locality sensitive hashing (BPLSH) indexing. Compared with existing Link-based Similarity join (LS-join) method, PS-join can derive various similarity semantics. Experimental results on real data sets show the efficiency and effectiveness of the proposed approach.

Description: Time-series classification has attracted considerable research attention due to the various domains where time-series data are observed, ranging from medicine to econometrics. Traditionally, the focus of time-series classification has been on short time-series data composed of a few patterns exhibiting variabilities, while recently there have been attempts to focus on longer series composed of multiple local patrepeating with an arbitrary irregularity. The primary contribution of this paper relies on presenting a method which can detect local patterns in repetitive time-series via fitting local polynomial functions of a specified degree. We capture the repetitiveness degrees of time-series datasets via a new measure. Furthermore, our method approximates local polynomials in linear time and ensures an overall linear running time complexity. The coefficients of the polynomial functions are converted to symbolic words via equi-area discretizations of the coefficients’ distributions. The symbolic polynomial words enable the detection of similar local patterns by assigning the same word to similar polynomials. Moreover, a histogram of the frequencies of the words is constructed from each time-series’ bag of words. Each row of the histogram enables a new representation for the series and symbolizes the occurrence of local patterns and their frequencies. In an experimental comparison against state-of-the-art baselines on repetitive datasets, our method demonstrates significant improvements in terms of prediction accuracy.

Description: Sentiment classification is a topic-sensitive task, i.e., a classifier trained from one topic will perform worse on another. This is especially a problem for the tweets sentiment analysis. Since the topics in Twitter are very diverse, it is impossible to train a universal classifier for all topics. Moreover, compared to product review, Twitter lacks data labeling and a rating mechanism to acquire sentiment labels. The extremely sparse text of tweets also brings down the performance of a sentiment classifier. In this paper, we propose a semi-supervised topic-adaptive sentiment classification (TASC) model, which starts with a classifier built on common features and mixed labeled data from various topics. It minimizes the hinge loss to adapt to unlabeled data and features including topic-related sentiment words, authors’ sentiments and sentiment connections derived from “@” mentions of tweets, named as topic-adaptive features. Text and non-text features are extracted and naturally split into two views for co-training. The TASC learning algorithm updates topic-adaptive features based on the collaborative selection of unlabeled data, which in turn helps to select more reliable tweets to boost the performance. We also design the adapting model along a timeline (TASC-t) for dynamic tweets. An experiment on 6 topics from published tweet corpuses demonstrates that TASC outperforms other well-known supervised and ensemble classifiers. It also beats those semi-supervised learning methods without feature adaption. Meanwhile, TASC-t can also achieve impressive accuracy and F-score. Finally, with timeline visualization of “river” graph, people can intuitively grasp the ups and downs of sentiments’ evolvement, and the intensity by color gradation.

Description: This paper presents a new robust EM algorithm for the finite mixture learning procedures. The proposed Spatial-EM algorithm utilizes median-based location and rank-based scatter estimators to replace sample mean and sample covariance matrix in each M step, hence enhancing stability and robustness of the algorithm. It is robust to outliers and initial values. Compared with many robust mixture learning methods, the Spatial-EM has the advantages of simplicity in implementation and statistical efficiency. We apply Spatial-EM to supervised and unsupervised learning scenarios. More specifically, robust clustering and outlier detection methods based on Spatial-EM have been proposed. We apply the outlier detection to taxonomic research on fish species novelty discovery. Two real datasets are used for clustering analysis. Compared with the regular EM and many other existing methods such as K-median, X-EM and SVM, our method demonstrates superior performance and high robustness.

Description: It is a big challenge to guarantee the quality of discovered relevance features in text documents for describing user preferences because of large scale terms and data patterns. Most existing popular text mining and classification methods have adopted term-based approaches. However, they have all suffered from the problems of polysemy and synonymy. Over the years, there has been often held the hypothesis that pattern-based methods should perform better than term-based ones in describing user preferences; yet, how to effectively use large scale patterns remains a hard problem in text mining. To make a breakthrough in this challenging issue, this paper presents an innovative model for relevance feature discovery. It discovers both positive and negative patterns in text documents as higher level features and deploys them over low-level features (terms). It also classifies terms into categories and updates term weights based on their specificity and their distributions in patterns. Substantial experiments using this model on RCV1, TREC topics and Reuters-21578 show that the proposed model significantly outperforms both the state-of-the-art term-based methods and the pattern based methods.

Description: Numerous theories and algorithms have been developed to solve vectorial data learning problems by searching for the hypothesis that best fits the observed training sample. However, many real-world applications involve samples that are not described as feature vectors, but as (dis)similarity data. Converting vectorial data into (dis)similarity data is more easily performed than converting (dis)similarity data into vectorial data. This study proposes a stochastic iterative distance transformation model for similarity-based learning. The proposed model can be used to identify a clear class boundary in data by modifying the (dis)similarities between examples. The experimental results indicate that the performance of the proposed method is comparable with those of various vector-based and proximity-based learning algorithms.

Description: Although graph embedding has been a powerful tool for modeling data intrinsic structures, simply employing all features for data structure discovery may result in noise amplification. This is particularly severe for high dimensional data with small samples. To meet this challenge, this paper proposes a novel efficient framework to perform feature selection for graph embedding, in which a category of graph embedding methods is cast as a least squares regression problem. In this framework, a binary feature selector is introduced to naturally handle the feature cardinality in the least squares formulation. The resultant integral programming problem is then relaxed into a convex Quadratically Constrained Quadratic Program (QCQP) learning problem, which can be efficiently solved via a sequence of accelerated proximal gradient (APG) methods. Since each APG optimization is w.r.t. only a subset of features, the proposed method is fast and memory efficient. The proposed framework is applied to several graph embedding learning problems, including supervised, unsupervised, and semi-supervised graph embedding. Experimental results on several high dimensional data demonstrated that the proposed method outperformed the considered state-of-the-art methods.

Description: Boosting is an iterative process that improves the predictive accuracy for supervised (machine) learning algorithms. Boosting operates by learning multiple functions with subsequent functions focusing on incorrect instances where the previous functions predicted the wrong label. Despite considerable success, boosting still has difficulty on data sets with certain types of problematic training data (e.g., label noise) and when complex functions overfit the training data. We propose a novel cluster-based boosting (CBB) approach to address limitations in boosting for supervised learning systems. Our CBB approach partitions the training data into clusters containing highly similar member data and integrates these clusters directly into the boosting process. CBB boosts selectively (using a high learning rate, low learning rate, or not boosting) on each cluster based on both the additional structure provided by the cluster and previous function accuracy on the member data. Selective boosting allows CBB to improve predictive accuracy on problematic training data. In addition, boosting separately on clusters reduces function complexity to mitigate overfitting. We provide comprehensive experimental results on 20 UCI benchmark data sets with three different kinds of supervised learning systems. These results demonstrate the effectiveness of our CBB approach compared to a popular boosting algorithm, an algorithm that uses clusters to improve boosting, and two algorithms that use selective boosting without clustering.

Description: Discovering trajectory patterns is shown to be very useful in learning interactions between moving objects. Many types of trajectory patterns have been proposed in the literature, but previous methods were developed for only a specific type of trajectory patterns. This limitation could make pattern discovery tedious and inefficient since users typically do not know which types of trajectory patterns are hidden in their data sets. Our main observation is that many trajectory patterns can be arranged according to the strength of temporal constraints. In this paper, we propose a unifying framework of mining trajectory patterns of various temporal tightness, which we call unifying trajectory patterns  ( UT-patterns ). This framework consists of two phases: initial pattern discovery and granularity adjustment . A set of initial patterns are discovered in the first phase, and their granularities (i.e., levels of detail) are adjusted by split and merge to detect other types in the second phase. As a result, the structure called a pattern forest is constructed to show various patterns. Both phases are guided by an information-theoretic formula without user intervention. Experimental results demonstrate that our framework facilitates easy discovery of various patterns from real-world trajectory data.

Description: With the increasing availability of graph data and widely adopted cloud computing paradigm, graph partitioning has become an efficient pre-processing technique to balance the computing workload and cope with the large scale of input data. Since the cost of partitioning the entire graph is strictly prohibitive, there are some recent tentative works towards streaming graph partitioning which run faster, are easily parallelized, and can be incrementally updated. Most of the existing works on streaming partitioning assume that worker nodes within a cluster are homogeneous in nature. Unfortunately, this assumption does not always hold. Experiments show that these homogeneous algorithms suffer a significant performance degradation when running at heterogeneous environment. In this paper, we propose a novel adaptive streaming graph partitioning approach to cope with heterogeneous environment. We first formally model the heterogeneous computing environment with the consideration of the unbalance of computing ability (e.g., the CPU frequency) and communication ability (e.g., the network bandwidth) for each node. Based on this model, we propose a new graph partitioning objective function that aims to minimize the total execution time of the graph-processing job. We then explore some simple yet effective streaming algorithms for this objective function that can achieve balanced and efficient partitioning result. Extensive experiments are conducted on a moderate sized computing cluster with real-world web and social network graphs. The results demonstrate that the proposed approach achieves significant improvement compared with the state-of-the-art solutions.

Description: Matrix factorization is useful to extract the essential low-rank structure from a given matrix and has been paid increasing attention. A typical example is non-negative matrix factorization (NMF), which is one type of unsupervised learning, having been successfully applied to a variety of data including documents, images and gene expression, where their values are usually non-negative. We propose a new model of NMF which is trained by using auxiliary information of overlapping groups. This setting is very reasonable in many applications, a typical example being gene function estimation where functional gene groups are heavily overlapped with each other. To estimate true groups from given overlapping groups efficiently, our model incorporates latent matrices with the regularization term using a mixed norm. This regularization term allows group-wise sparsity on the optimized low-rank structure. The latent matrices and other parameters are efficiently estimated by a block coordinate gradient descent method. We empirically evaluated the performance of our proposed model and algorithm from a variety of viewpoints, comparing with four methods including MMF for auxiliary graph information, by using both synthetic and real world document and gene expression data sets.

Description: This paper focuses on the measure of recommendation stability, which reflects the consistency of recommender system predictions. Stability is a desired property of recommendation algorithms and has important implications on users’ trust and acceptance of recommendations. Prior research has reported that some popular recommendation algorithms can suffer from a high degree of instability. In this study, we explore two scalable, general-purpose meta-algorithmic approaches—based on bagging and iterative smoothing—that can be used in conjunction with different traditional recommendation algorithms to improve their stability. Our experimental results on real-world rating data demonstrate that both approaches can achieve substantially higher stability as compared to the original recommendation algorithms. Furthermore, perhaps as importantly, the proposed approaches not only do not sacrifice the predictive accuracy in order to improve recommendation stability, but are actually able to provide additional accuracy improvements.

Description: We propose selective supervised Latent Dirichlet Allocation (ssLDA) to boost the prediction performance of the widely studied supervised probabilistic topic models. We introduce a Bernoulli distribution for each word in one given document to select this word as a strongly or weakly discriminative one with respect to its assigned topic. The Bernoulli distribution is parameterized by the discrimination power of the word for its assigned topic. As a result, the document is represented as a “bag-of-selective-words” instead of the probabilistic “bag-of-topics” in the topic modeling domain or the flat “bag-of-words” in the traditional natural language processing domain to form a new perspective. Inheriting the general framework of supervised LDA (sLDA), ssLDA can also predict many types of response specified by a Gaussian Linear Model (GLM). Focusing on the utilization of this word selection mechanism for singe-label document classification in this paper, we conduct the variational inference for approximating the intractable posterior and derive a maximum-likelihood estimation of parameters in ssLDA. The experiments reported on textual documents show that ssLDA not only performs competitively over “state-of-the-art” classification approaches based on both the flat “bag-of-words” and probabilistic “bag-of-topics” representation in terms of classification performance, but also has the ability to discover the discrimination power of the words specified in the topics (compatible with our rational knowledge).

Description: Metric learning, the task of learning a good distance metric, is a key problem in machine learning with ample applications. This paper introduces a novel framework for nonlinear metric learning, called kernel density metric learning (KDML), which is easy to use and provides nonlinear, probability-based distance measures. KDML constructs a direct nonlinear mapping from the original input space into a feature space based on kernel density estimation. The nonlinear mapping in KDML embodies established distance measures between probability density functions, and leads to accurate classification on datasets for which existing linear metric learning methods would fail. It addresses the severe challenge to distance-based classifiers when features are from heterogeneous domains and, as a result, the Euclidean or Mahalanobis distance between original feature vectors is not meaningful. We also propose two ways to determine the kernel bandwidths, including an adaptive local scaling approach and an integrated optimization algorithm that learns the Mahalanobis matrix and kernel bandwidths together. KDML is a general framework that can be combined with any existing metric learning algorithm. As concrete examples, we combine KDML with two leading metric learning algorithms, large margin nearest neighbors (LMNN) and neighborhood component analysis (NCA). KDML can naturally handle not only numerical features, but also categorical ones, which is rarely found in previous metric learning algorithms. Extensive experimental results on various datasets show that KDML significantly improves existing metric learning algorithms in terms of classification accuracy.

Description: The Earth Mover’s Distance (EMD) is a well-known distance metric for data represented as probability distributions over a predefined feature space. Supporting EMD-based similarity search has attracted intensive research effort. Despite the plethora of literature, most existing solutions are optimized for $L^p$ feature spaces (e.g., Euclidean space); while in a spectrum of applications, the relationships between features are better captured using networks. In this paper, we study the problem of answering $k$ -nearest neighbor ( $k$ -NN) queries under network-based EMD metrics (NEMD). We propose Oasis , a new access method which leverages the network structure of feature space and enables efficient NEMD-based similarity search. Specifically, Oasis employs three novel techniques: (i) Range Oracle , a scalable model to estimate the range of $k$ -th nearest neighbor under NEMD, (ii) Boundary Index , a structure that efficiently fetches candidates within given range, and (iii) Network Compression Hierarchy , an incremental filtering mechanism that effectively prunes false positive candidates to save unnecessary computation. Through extensive experiments using both synthetic and real data sets, we confirmed that Oasis significantly outperforms the state-of-the-art methods in query processing cost.

Description: Many mature term-based or pattern-based approaches have been used in the field of information filtering to generate users’ information needs from a collection of documents. A fundamental assumption for these approaches is that the documents in the collection are all about one topic. However, in reality users’ interests can be diverse and the documents in the collection often involve multiple topics. Topic modelling, such as Latent Dirichlet Allocation (LDA), was proposed to generate statistical models to represent multiple topics in a collection of documents, and this has been widely utilized in the fields of machine learning and information retrieval, etc. But its effectiveness in information filtering has not been so well explored. Patterns are always thought to be more discriminative than single terms for describing documents. However, the enormous amount of discovered patterns hinder them from being effectively and efficiently used in real applications, therefore, selection of the most discriminative and representative patterns from the huge amount of discovered patterns becomes crucial. To deal with the above mentioned limitations and problems, in this paper, a novel information filtering model, Maximum matched Pattern-based Topic Model (MPBTM), is proposed. The main distinctive features of the proposed model include: (1) user information needs are generated in terms of multiple topics; (2) each topic is represented by patterns; (3) patterns are generated from topic models and are organized in terms of their statistical and taxonomic features; and (4) the most discriminative and representative patterns, called Maximum Matched Patterns, are proposed to estimate the document relevance to the user’s information needs in order to filter out irrelevant documents. Extensive experiments are conducted to evaluate the effectiveness of the proposed model by using the TREC data collection Reuters Corpus Volume 1. The results show that the proposed- model significantly outperforms both state-of-the-art term-based models and pattern-based models.

Description: Trust plays a crucial role in helping online users collect reliable information and it has gained increasing attention from the computer science community in recent years. Traditionally, research about online trust assumes static trust relations between users. However, trust, as a social concept, evolves as people interact. Most existing studies about trust evolution are from sociologists in the physical world while little work exists in an online world. Studying online trust evolution faces unique challenges because more often than not, available data is from passive observation. In this work, we leverage social science theories to develop a methodology that enables the study of online trust evolution. In particular, we identify the differences of trust evolution study in physical and online worlds and propose a framework, eTrust, to study trust evolution using online data from passive observation in the context of product review sites by exploiting the dynamics of user preferences. We present technical details about modeling trust evolution, and perform experiments to show how the exploitation of trust evolution can help improve the performance of online applications such as trust prediction, rating prediction and ranking evolution.

Description: Learning to rank arises in many data mining applications, ranging from web search engine, online advertising to recommendation system. In learning to rank, the performance of a ranking model is strongly affected by the number of labeled examples in the training set; on the other hand, obtaining labeled examples for training data is very expensive and time-consuming. This presents a great need for the active learning approaches to select most informative examples for ranking learning; however, in the literature there is still very limited work to address active learning for ranking. In this paper, we propose a general active learning framework, expected loss optimization (ELO), for ranking. The ELO framework is applicable to a wide range of ranking functions. Under this framework, we derive a novel algorithm, expected discounted cumulative gain (DCG) loss optimization (ELO-DCG), to select most informative examples. Then, we investigate both query and document level active learning for raking and propose a two-stage ELO-DCG algorithm which incorporate both query and document selection into active learning. Furthermore, we show that it is flexible for the algorithm to deal with the skewed grade distribution problem with the modification of the loss function. Extensive experiments on real-world web search data sets have demonstrated great potential and effectiveness of the proposed framework and algorithms.

Description: Advanced technology in GPS and sensors enables us to track physical events, such as human movements and facility usage. Periodicity analysis from the recorded data is an important data mining task which provides useful insights into the physical events and enables us to report outliers and predict future behaviors. To mine periodicity in an event, we have to face real-world challenges of inherently complicated periodic behaviors and imperfect data collection problem. Specifically, the hidden temporal periodic behaviors could be oscillating and noisy, and the observations of the event could be incomplete. In this paper, we propose a novel probabilistic measure for periodicity and design a practical algorithm, ePeriodicity, to detect periods. Our method has thoroughly considered the uncertainties and noises in periodic behaviors and is provably robust to incomplete observations. Comprehensive experiments on both synthetic and real datasets demonstrate the effectiveness of our method.

Description: Duplicate detection is the process of identifying multiple representations of same real world entities. Today, duplicate detection methods need to process ever larger datasets in ever shorter time: maintaining the quality of a dataset becomes increasingly difficult. We present two novel, progressive duplicate detection algorithms that significantly increase the efficiency of finding duplicates if the execution time is limited: They maximize the gain of the overall process within the time available by reporting most results much earlier than traditional approaches. Comprehensive experiments show that our progressive algorithms can double the efficiency over time of traditional duplicate detection and significantly improve upon related work.

Description: The classification of patterns into naturally ordered labels is referred to as ordinal regression or ordinal classification. Usually, this classification setting is by nature highly imbalanced, because there are classes in the problem that are a priori more probable than others. Although standard over-sampling methods can improve the classification of minority classes in ordinal classification, they tend to introduce severe errors in terms of the ordinal label scale, given that they do not take the ordering into account. A specific ordinal over-sampling method is developed in this paper for the first time in order to improve the performance of machine learning classifiers. The method proposed includes ordinal information by approaching over-sampling from a graph-based perspective. The results presented in this paper show the good synergy of a popular ordinal regression method (a reformulation of support vector machines) with the graph-based proposed algorithms, and the possibility of improving both the classification and the ordering of minority classes. A cost-sensitive version of the ordinal regression method is also introduced and compared with the over-sampling proposals, showing in general lower performance for minority classes.

Description: This paper presents a novel approach that transforms the feature space into a new feature space such that a range query in the original space is mapped into an equivalent box query in the transformed space. Since box queries are axis aligned, there are several implementational advantages that can be exploited to speed up the retrieval of query results using R-Tree [9] like indexing schemes. For two dimensional data, the transformation is precise. For larger than two dimensions, we propose a space transformation scheme based on disjoint planer rotation and a new type of query, pruning box query, to get the precise results. Experimental results with large synthetic databases and some real databases show the effectiveness of the proposed transformation scheme. These experimental results have been corroborated with suitable mathematical models. In disjoint planer rotation, additional computation time is required to remove the false positives produced due to the bounding box not being precise. A second topological transformation scheme is presented based on optimized bounding box, which reduces the amount of false positives. The amount of this reduction is more with increasing dimensions. Optimized bounding box for higher dimensions is computed based on a novel approach of simultaneous local optimal projections.

Description: In a relational database, tuples are called “duplicate” if they describe the same real-world entity. If such duplicate tuples are observed, it is recommended to remove them and to replace them with one tuple that represents the joint information of the duplicate tuples to a maximal extent. This remove-and-replace operation is called a fusion operation. Within the setting of a relational database management system, the removal of the original duplicate tuples can breach referential integrity. In this paper, a strategy is proposed to maintain referential integrity in a semantically correct manner, thereby optimizing the quality of relationships in the database. An algorithm is proposed that is able to propagate a fusion operation through the entire database. The algorithm is based on a framework of first and second order fusion functions on the one hand, and conflict resolution strategies on the other hand. It is shown how classical strategies for maintaining referential integrity, such as delete cascading, are highly specialized cases of the proposed framework. Experimental results are reported that (i) show the efficiency of the proposed algorithm and (ii) show the differences in quality between several second order fusion functions. It is shown that some strategies easily outperform delete cascading.

Description: Rapidly escalating energy and cooling costs, especially those related to the energy consumption of storage systems, have become a concern for data centers, primarily because the amount of digital data that needs storage is increasing daily. In response, a multitude of energy saving approaches that take into account storage-device-level input/output (I/O) behaviors have been proposed. The trouble is that numerous critical applications such as database systems or web commerce applications are in constant operation at data centers, and the conventional approaches that only utilize storage-device-level I/O behaviors do not produce sufficient energy savings. It may be possible to dramatically reduce storage-related energy consumption without degrading application performance levels by utilizing application-level I/O behaviors. However, such behaviors differ from one application to another, and it would be too expensive to tailor methods to individual applications. As a way of solving this problem, we propose a universal storage energy management framework for runtime storage energy savings that can be applied to any type of application. The results of evaluations show that the use of this framework results in substantive energy savings compared with the traditional approaches that are used while applications are running.

Description: The data deduplication task has attracted a considerable amount of attention from the research community in order to provide effective and efficient solutions. The information provided by the user to tune the deduplication process is usually represented by a set of manually labeled pairs. In very large datasets, producing this kind of labeled set is a daunting task since it requires an expert to select and label a large number of informative pairs. In this article, we propose a two-stage sampling selection strategy (T3S) that selects a reduced set of pairs to tune the deduplication process in large datasets. T3S selects the most representative pairs by following two stages. In the first stage, we propose a strategy to produce balanced subsets of candidate pairs for labeling. In the second stage, an active selection is incrementally invoked to remove the redundant pairs in the subsets created in the first stage in order to produce an even smaller and more informative training set. This training set is effectively used both to identify where the most ambiguous pairs lie and to configure the classification approaches. Our evaluation shows that T3S is able to reduce the labeling effort substantially while achieving a competitive or superior matching quality when compared with state-of-the-art deduplication methods in large datasets.

Description: Graph datasets with billions of edges, such as social and web graphs, are prevalent, and scalability is critical. All-distances sketches (ADS) [Cohen 1997], are a powerful tool for scalable approximation of statistics. The sketch is a small size sample of the distance relation of a node which emphasizes closer nodes. Sketches for all nodes are computed using a nearly linear computation and estimators are applied to sketches of nodes to estimate their properties. We provide, for the first time, a unified exposition of ADS algorithms and applications. We present the historic inverse probability (HIP) estimators which are applied to the ADS of a node to estimate a large natural class of statistics. For the important special cases of neighborhood cardinalities (the number of nodes within some query distance) and closeness centralities, HIP estimators have at most half the variance of previous estimators and we show that this is essentially optimal. Moreover, HIP obtains a polynomial improvement for more general statistics and the estimators are simple, flexible, unbiased, and elegant. For approximate distinct counting on data streams, HIP outperforms the original estimators for the HyperLogLog MinHash sketches (Flajolet et al. 2007), obtaining significantly improved estimation quality for this state-of-the-art practical algorithm.

Description: In emerging information networks, it is crucially important to provide efficient search on distributed documents while preserving their owners’ privacy, for which privacy preserving indexes or PPI presents a possible solution. An understudied problem for the PPI techniques is how to provide differentiated privacy preservation in the presence of multi-keyword document search. The differentiation is necessary as terms and phrases bear innate differences in their semantic meanings. In this paper, we present $epsilon$ -mPPI , the first work to provide the distributed document search with quantitatively differentiated privacy preservation. In the design of $epsilon$ -mPPI , we identified a suite of challenging problems and proposed novel solutions. For one, we formulated the quantitative privacy computation as an optimization problem that strikes a balance between privacy preservation and search efficiency. We also addressed the challenging problem of secure $epsilon$ -mPPI construction in the multi-domain information network which lacks mutual trusts between domains. Towards a secure $epsilon$ -mPPI construction with practically acceptable performance, we proposed to optimize the performance of secure multi-party computations by m- king a novel use of secret sharing. We implemented the $epsilon$ -mPPI construction protocol with a functioning prototype. We conducted extensive experiments to evaluate the prototype’s effectiveness and efficiency based on a real-world dataset.

Description: The current Web has no general mechanisms to make digital artifacts—such as datasets, code, texts, and images—verifiable and permanent. For digital artifacts that are supposed to be immutable, there is moreover no commonly accepted method to enforce this immutability. These shortcomings have a serious negative impact on the ability to reproduce the results of processes that rely on Web resources, which in turn heavily impacts areas such as science where reproducibility is important. To solve this problem, we propose trusty URIs containing cryptographic hash values. We show how trusty URIs can be used for the verification of digital artifacts, in a manner that is independent of the serialization format in the case of structured data files such as nanopublications. We demonstrate how the contents of these files become immutable, including dependencies to external digital artifacts and thereby extending the range of verifiability to the entire reference tree. Our approach sticks to the core principles of the Web, namely openness and decentralized architecture, and is fully compatible with existing standards and protocols. Evaluation of our reference implementations shows that these design goals are indeed accomplished by our approach, and that it remains practical even for very large files.

Description: In this paper, we present a novel transfer learning framework for network node classification. Our objective is to accurately predict the labels of nodes in a target network by leveraging information from an auxiliary source network. Such a transfer learning framework is potentially useful for broader areas of network classification, where emerging new networks might not have sufficient labeled information because node labels are either costly to obtain or simply not available, whereas many established networks from related domains are available to benefit the learning. In reality, the source and the target networks may not share common nodes or connections, so the major challenge of cross-network transfer learning is to identify knowledge/patterns transferable between networks and potentially useful to support cross-network learning. In this work, we propose to learn common signature subgraphs between networks, and use them to construct new structure features for the target network. By combining the original node content features and the new structure features, we develop an iterative classification algorithm, TrGraph, that utilizes label dependency to jointly classify nodes in the target network. Experiments on real-world networks demonstrate that TrGraph achieves the superior performance compared to the state-of-the-art baseline methods, and transferring generalizable structure information can indeed improve the node classification accuracy.

Description: Increasingly, aiming to contain their rapidly growing energy expenditures, commercial buildings are equipped to respond to utility’s demand and price signals. Such smart energy consumption, however, heavily relies on accurate short-term energy load forecasting, such as hourly predictions for the next $n$ $(nge 2)$ hours. To attain sufficient accuracy for these predictions, it is important to exploit the relationships among the $n$ estimated outputs. This paper treats such multi-steps ahead regression task as a sequence labeling (regression) problem, and adopts the continuous conditional random fields (CCRF) to explicitly model these interconnected outputs. In particular, we improve the CCRF’s computation complexity and predictive accuracy with two novel strategies. First, we employ two tridiagonal matrix computation techniques to significantly speed up the CCRF’s training and inference. These techniques tackle the cubic computational cost required by the matrix inversion calculations in the training and inference of the CCRF, resulting in linear complexity for these matrix operations. Second, we address the CCRF’s weak feature constraint problem with a novel multi-target edge function, thus boosting the CCRF’s predictive performance. The proposed multi-target feature is able to convert the relationship of related outputs with continuous values into a set of “sub-relationships”, each providing more specific feature constraints for the interplays of the r- lated outputs. We applied the proposed approach to two real-world energy load prediction systems: one for electricity demand and another for gas usage. Our experimental results show that the proposed strategy can meaningfully reduce the predictive error for the two systems, in terms of mean absolute percentage error and root mean square error, when compared with three benchmarking methods. Promisingly, the relative error reduction achieved by our CCRF model was up to 50 percent.

Description: Automatic disease inference is of importance to bridge the gap between what online health seekers with unusual symptoms need and what busy human doctors with biased expertise can offer. However, accurately and efficiently inferring diseases is non-trivial, especially for community-based health services due to the vocabulary gap, incomplete information, correlated medical concepts, and limited high quality training samples. In this paper, we first report a user study on the information needs of health seekers in terms of questions and then select those that ask for possible diseases of their manifested symptoms for further analytic. We next propose a novel deep learning scheme to infer the possible diseases given the questions of health seekers. The proposed scheme is comprised of two key components. The first globally mines the discriminant medical signatures from raw features. The second deems the raw features and their signatures as input nodes in one layer and hidden nodes in the subsequent layer, respectively. Meanwhile, it learns the inter-relations between these two layers via pre-training with pseudo-labeled data. Following that, the hidden nodes serve as raw features for the more abstract signature mining. With incremental and alternative repeating of these two components, our scheme builds a sparsely connected deep architecture with three hidden layers. Overall, it well fits specific tasks with fine-tuning. Extensive experiments on a real-world dataset labeled by online doctors show the significant performance gains of our scheme.

Description: Today, plenty of apps are released to enable users to make the best use of their cell phones. Facing the large amount of apps, app retrieval and app recommendation become important, since users can easily use them to acquire their desired apps. To obtain high-quality retrieval and recommending results, it needs to obtain the precise app relationship calculating results. Unfortunately, the recent methods are conducted mostly relying on user's log or app's description, which can only detect whether two apps are downloaded, installed meanwhile or provide similar functions or not. In fact, apps contain many general relationships other than similarity, such as one app needs another app as its tool. These relationships cannot be dug via user's log or app's description. Reviews contain user's viewpoint and judgment to apps, thus they can be used to calculate relationship between apps. To use reviews, this paper proposes an iterative process by combining review similarity and app relationship together. Experimental results demonstrate that via this iterative process, relationship between apps can be calculated exactly. Furthermore, this process is improved in two aspects. One is to obtain excellent results even with weak initialization. The other is to apply matrix product to reduce running time.

Description: Earth Mover’s Distance (EMD) evaluates the similarity between probability distributions, known as a robust measure more consistent with human similarity perception than traditional similarity functions. EMD-based similarity join retrieves pairs of probability distributions with EMD below a specified threshold, supporting many important applications, such as duplicate image retrieval and sensor pattern recognition. This paper studies the possibility of using MapReduce to improve the scalability of EMD similarity join. While existing MapReduce optimization techniques mainly aim to minimize the communication overhead, such methods are not applicable to our problem, due to the high computational cost of EMD. Utilizing the dual-program mapping technique, we present a new general data partition framework to facilitate effective workload decomposition using MapReduce, ensuring similar distributions in terms of EMD are mapped to the same reduce task for further verification. New optimization strategies are also proposed to balance the workloads among reduce tasks and eliminate large unnecessary EMD evaluations. Our experiments verify the superiority of our proposal on system efficiency, with a huge advantage of at least one order of magnitude than the state-of-the-art solution, and on system effectiveness, with a real case study towards the abused image phenomenon on the most popular C2C website in China.

Description: A large number of URLs collected by web crawlers correspond to pages with duplicate or near-duplicate contents. To crawl, store, and use such duplicated data implies a waste of resources, the building of low quality rankings, and poor user experiences. To deal with this problem, several studies have been proposed to detect and remove duplicate documents without fetching their contents. To accomplish this, the proposed methods learn normalization rules to transform all duplicate URLs into the same canonical form. A challenging aspect of this strategy is deriving a set of general and precise rules. In this work, we present DUSTER, a new approach to derive quality rules that take advantage of a multi-sequence alignment strategy. We demonstrate that a full multi-sequence alignment of URLs with duplicated content, before the generation of the rules, can lead to the deployment of very effective rules. By evaluating our method, we observed it achieved larger reductions in the number of duplicate URLs than our best baseline, with gains of 82 and 140.74 percent in two different web collections.

Description: A fundamental problem of graph databases is subgraph isomorphism query (a.k.a subgraph query): given a query graph $Q$ and a graph database, it retrieves the graphs $G$ s from the database that contain $Q$ . Due to the cost of managing massive data coupled with the computational hardness of subgraph isomorphism testing, outsourcing the computations to a third-party provider is an appealing alternative. However, confidentiality has been a critical attribute of quality of service ( QoS ) in query services. To the best of our knowledge, subgraph query services with tunable preservation of privacy of structural information have never been addressed. In this paper, we present the first work on structure-preserving $mathsf {subIso}$ ( $mathsf {SPsubIso}$ ). A crucial step of our work is to transform $mathsf {subIso}$ —the seminal subgraph isomorphism algorithm (the Ullmann’s algorithm)—into a series of matrix operations. We propose a novel cyclic- group based encryption ( $mathsf {CGBE}$ ) method for private matrix operations. We propose a protocol that involves the query client and static indexes to optimize $mathsf {SPsubIso}$ . We prove that the structural information of both $Q$ and $G$ are preserved under $mathsf {CGBE}$ and analyze the privacy preservation in the presence of the optimizations. Our extensive experiments on both real and synthetic datasets verify that $mathsf {SPsubIso}$ is efficient and the optimizations are effective.

Description: Existing main-memory hash join algorithms for multi-core can be classified into two camps. Hardware-oblivious hash join variants do not depend on hardware-specific parameters. Rather, they consider qualitative characteristics of modern hardware and are expected to achieve good performance on any technologically similar platform. The assumption behind these algorithms is that hardware is now good enough at hiding its own limitations—through automatic hardware prefetching, out-of-order execution, or simultaneous multi-threading (SMT)—to make hardware-oblivious algorithms competitive without the overhead of carefully tuning to the underlying hardware. Hardware-conscious implementations, such as (parallel) radix join , aim to maximally exploit a given architecture by tuning the algorithm parameters (e.g., hash table sizes) to the particular features of the architecture. The assumption here is that explicit parameter tuning yields enough performance advantages to warrant the effort required. This paper compares the two approaches under a wide range of workloads (relative table sizes, tuple sizes, effects of sorted data, etc.) and configuration parameters (VM page sizes, number of threads, number of cores, SMT, SIMD, prefetching, etc.). The results show that hardware-conscious algorithms generally outperform hardware-oblivious ones. However, on specific workloads and special architectures with aggressive simultaneous multi-threading, hardware-oblivious algorithms are competitive. The main conclusion of the paper is that, in existing multi-core architectures, it is still important to carefully tailor algorithms to the underlying hardware to get the necessary performance. But processor developments may require to revisit this conclusion in the future.

Description: Hybrid human/computer database systems promise to greatly expand the usefulness of query processing by incorporating the crowd for data gathering and other tasks. Such systems raise many implementation questions. Perhaps the most fundamental issue is that the closed world assumption underlying relational query semantics does not hold in such systems. As a consequence, the meaning of even simple queries can be called into question. Furthermore, query progress monitoring becomes difficult due to non-uniformities in the arrival of crowdsourced data and peculiarities of how people work in crowdsourcing systems. To address these issues, we develop statistical tools that enable users and systems developers to reason about query completeness. These tools can also help drive query execution and crowdsourcing strategies. We evaluate our techniques using experiments on a popular crowdsourcing platform.

Description: Recently, there has been a growing interest in designing differentially private data mining algorithms. Frequent itemset mining (FIM) is one of the most fundamental problems in data mining. In this paper, we explore the possibility of designing a differentially private FIM algorithm which can not only achieve high data utility and a high degree of privacy, but also offer high time efficiency. To this end, we propose a differentially private FIM algorithm based on the FP-growth algorithm, which is referred to as PFP-growth. The PFP-growth algorithm consists of a preprocessing phase and a mining phase. In the preprocessing phase, to improve the utility and privacy tradeoff, a novel smart splitting method is proposed to transform the database. For a given database, the preprocessing phase needs to be performed only once. In the mining phase, to offset the information loss caused by transaction splitting, we devise a run-time estimation method to estimate the actual support of itemsets in the original database. In addition, by leveraging the downward closure property, we put forward a dynamic reduction method to dynamically reduce the amount of noise added to guarantee privacy during the mining process. Through formal privacy analysis, we show that our PFP-growth algorithm is $epsilon$ -differentially private. Extensive experiments on real datasets illustrate that our PFP-growth algorithm substantially outperforms the state-of-the-art techniques.

Description: SimRank is a powerful model for assessing vertex-pair similarities in a graph. It follows the concept that two vertices are similar if they are referenced by similar vertices. The prior work [18] exploits partial sums memoization to compute SimRank in $O(Kmn)$ time on a graph of $n$ vertices and $m$ edges, for $K$ iterations. However, computations among different partial sums may have redundancy. Besides, to guarantee a given accuracy $epsilon$ , the existing SimRank needs $K=lceil log _C ,epsilon rceil$ iterations, where $C$ is a damping factor, but the geometric rate of convergence is slow if a high accuracy is expected. In this paper, (1) a novel clustering strategy is proposed to eliminate duplicate computations occurring in partial sums, - nd an efficient algorithm is then devised to accelerate SimRank computation to $O(K d^{prime } n^2)$ time, where $d^{prime }$ is typically much smaller than $frac{m}{n}$ . (2) A new differential SimRank equation is proposed, which can represent the SimRank matrix as an exponential sum of transition matrices, as opposed to the geometric sum of the conventional counterpart. This leads to a further speedup in the convergence rate of SimRank iterations. (3) In bipartite domains, a novel finer-grained partial max clustering method is developed to speed up the computation of the Minimax SimRank variation from $O(Kmn)$ to $O(Km^{prime }n)$ time, where $m^{prime } ({le} m)$ is the number of edges in a reduced graph after edge clustering, which can be typically much smaller than $m$ . Usi

Description: Growing main memory capacity has fueled the development of in-memory big data management and processing. By eliminating disk I/O bottleneck, it is now possible to support interactive data analytics. However, in-memory systems are much more sensitive to other sources of overhead that do not matter in traditional I/O-bounded disk-based systems. Some issues such as fault-tolerance and consistency are also more challenging to handle in in-memory environment. We are witnessing a revolution in the design of database systems that exploits main memory as its data storage layer. Many of these researches have focused along several dimensions: modern CPU and memory hierarchy utilization, time/space efficiency, parallelism, and concurrency control. In this survey, we aim to provide a thorough review of a wide range of in-memory data management and processing proposals and systems, including both data storage systems and data processing frameworks. We also give a comprehensive presentation of important technology in memory management, and some key factors that need to be considered in order to achieve efficient in-memory data management and processing.

Description: Geo-textual data are generated in abundance. Recent studies focused on the processing of spatial keyword queries which retrieve objects that match certain keywords within a spatial region. To ensure effective retrieval, various extensions were done including the allowance of errors in keyword matching and autocompletion using prefix matching. In this paper, we propose INSPIRE, a general framework, which adopts a unifying strategy for processing different variants of spatial keyword queries. We adopt the autocompletion paradigm that generates an initial query as a prefix matching query. If there are few matching results, other variants are performed as a form of relaxation that reuses the processing done in the earlier phase. The types of relaxation allowed include spatial region expansion and exact/approximate prefix/substring matching. Moreover, since the autocompletion paradigm allows appending characters after the initial query, we look at how query processing done for the initial query and relaxation can be reused in such instances. Compared to existing works which process variants of spatial keyword query as new queries over different indexes, our approach offers a more compelling way to efficient and effective spatial keyword search. Extensive experiments substantiate our claims.

Description: Social identity linkage across different social media platforms is of critical importance to business intelligence by gaining from social data a deeper understanding and more accurate profiling of users. In this paper, we propose a solution framework, $sf{ HYDRA}$ , which consists of three key steps: (I) we model heterogeneous behavior by long-term topical distribution analysis and multi-resolution temporal behavior matching against high noise and information missing, and the behavior similarity are described by multi-dimensional similarity vector for each user pair; (II) we build structure consistency models to maximize the structure and behavior consistency on users’ core social structure across different platforms, thus the task of identity linkage can be performed on groups of users, which is beyond the individual level linkage in previous study; and (III) we propose a normalized-margin-based linkage function formulation, and learn the linkage function by multi-objective optimization where both supervised pair-wise linkage function learning and structure consistency maximization are conducted towards a unified Pareto optimal solution. The model is able to deal with drastic information missing, and avoid the curse-of-dimensionality in handling high dimensional sparse representation. Extensive experiments on 10 million users across seven popular social networks platforms demonstrate that $sf{ HYDRA}$ correctly identifies real user linkage across different platforms from massive noisy user behavior data records, and outperforms existing state-of-the-art approaches by at least 20 percent under differen- settings, and four times better in most settings.

Description: In this paper, we introduce a new max-margin discriminant projection method, which takes advantage of the latent variable representation for support vector machine (SVM) as the classification criterion. Specifically, the proposed model jointly learns the discriminative subspace and classifier in a Bayesian framework by conditioning on augmented variables. Moreover, an extended nonlinear model is developed based on the kernel trick, where the similar model can be used in this setting with few modifications. To explore the sparsity in the kernel expansion, we use the spike-and-slab prior to seek basis vectors (BVs) from the corresponding candidates. Unlike existing methods, which employ BVs to approximate the original feature space, in our method BVs are sought to associate the final classification task. Thanks to the conditionally conjugate property, the parameters in our models can be inferred via the simple and efficient Gibbs sampler. Finally, we test our methods on synthesized and real-world data, including large-scale data sets to demonstrate their efficiency and effectiveness.

Description: Access control mechanisms and Privacy Protection Mechanisms (PPM) have been proposed for data streams. The access control for a data stream allows roles access to tuples satisfying an authorized predicate sliding-window query. Sharing the sensitive stream data without PPM can compromise the privacy. The PPM meets privacy requirements, e.g., $k$ -anonymity or $l$ -diversity by generalization of stream data. Imprecision introduced by generalization can be reduced by delaying the publishing of stream data. However, the delay in sharing the stream tuples to achieve better accuracy can lead to false-negatives if the tuples are held by PPM while the query predicate is evaluated. Administrator of an access control policy defines the imprecision bound for each query. The challenge for PPM is to optimize the delay in publishing of stream data so that the imprecision bound for the maximum number of queries is satisfied. We formulate the precision-bounded access control for privacy-preserving data streams problem, present the hardness results, provide an anonymization algorithm, and conduct experimental evaluation of the proposed algorithm. Experiments demonstrate that the proposed heuristic provides better precision for a given data stream access control policy as compared to the minimum or maximum delay heuristics proposed in existing literature.

Description: Link prediction is one of the most fundamental problems in graph modeling and mining. It has been studied in a wide range of scenarios, from uncovering missing links between different entities in databases, to recommending relations between people in social networks. In this problem, we wish to predict unseen links in a growing target network by exploiting existing structures in source networks. Most of the existing methods often assume that abundant links are available in the target network to build a model for link prediction. However, in many scenarios, the target network may be too sparse to enable robust inference process, which makes link prediction challenging with the paucity of link data. On the other hand, in many cases, other (more densely linked) auxiliary networks can be available that contains similar link structure relevant to that in the target network. The linkage information in the existing networks can be used in conjunction with the node attribute information in both networks in order to make more accurate link recommendations. Thus, this paper proposes the use of learning methods to perform link inference by transferring the link information from the source network to the target network. We also note that the source network may contain the link information irrelevant to the target network. This leads to cross-network bias between the networks, which makes the link model built upon the source network misaligned with the link structure of the target network. Therefore, we re-sample the source network to rectify such cross-network bias by maximizing the cross-network relevance measured by the node attributes, as well as preserving as rich link information as possible to avoid the loss of source link structure caused by the re-sampling algorithm. The link model based on the re-sampled source network can make more accurate link predictions on the target network with aligned link structures across the networks. We present experimenta- results illustrating the effectiveness of the approach.

Description: We proposed and developed a taxi-sharing system that accepts taxi passengers’ real-time ride requests sent from smartphones and schedules proper taxis to pick up them via ridesharing, subject to time, capacity, and monetary constraints. The monetary constraints provide incentives for both passengers and taxi drivers: passengers will not pay more compared with no ridesharing and get compensated if their travel time is lengthened due to ridesharing; taxi drivers will make money for all the detour distance due to ridesharing. While such a system is of significant social and environmental benefit, e.g., saving energy consumption and satisfying people's commute, real-time taxi-sharing has not been well studied yet. To this end, we devise a mobile-cloud architecture based taxi-sharing system. Taxi riders and taxi drivers use the taxi-sharing service provided by the system via a smart phone App. The Cloud first finds candidate taxis quickly for a taxi ride request using a taxi searching algorithm supported by a spatio-temporal index. A scheduling process is then performed in the cloud to select a taxi that satisfies the request with minimum increase in travel distance. We built an experimental platform using the GPS trajectories generated by over 33,000 taxis over a period of three months. A ride request generator is developed (available at http://cs.uic.edu/∼sma/ridesharing) in terms of the stochastic process modelling real ride requests learned from the data set. Tested on this platform with extensive experiments, our proposed system demonstrated its efficiency, effectiveness and scalability. For example, when the ratio of the number of ride requests to the number of taxis is 6, our proposed system serves three times as many taxi riders as that when no ridesharing is performed while saving 11 percent in total travel distance and 7 percent taxi fare per rider.

Description: With the rise of social networks, large-scale graph analysis becomes increasingly important. Because SQL lacks the expressiveness and performance needed for graph algorithms, lower-level, general-purpose languages are often used instead. For greater ease of use and efficiency, we propose SociaLite, a high-level graph query language based on Datalog. As a logic programming language, Datalog allows many graph algorithms to be expressed succinctly. However, its performance has not been competitive when compared to low-level languages. With SociaLite, users can provide high-level hints on the data layout and evaluation order; they can also define recursive aggregate functions which, as long as they are meet operations, can be evaluated incrementally and efficiently. Moreover, recursive aggregate functions make it possible to implement more graph algorithms that cannot be implemented in Datalog. We evaluated SociaLite by running nine graph algorithms in total; eight for social network analysis (shortest paths, PageRank, hubs and authorities, mutual neighbors, connected components, triangles, clustering coefficients, and betweenness centrality) and one for biological network analysis (Eulerian cycles). We use two real-life social graphs, LiveJournal and Last.fm, for the evaluation as well as one synthetic graph. The optimizations proposed in this paper speed up almost all the algorithms by 3 to 22 times. SociaLite even outperforms typical Java implementations by an average of 50 percent for the graph algorithms tested. When compared to highly optimized Java implementations, SociaLite programs are an order of magnitude more succinct and easier to write. Its performance is competitive, with only 16 percent overhead for the largest benchmark, and 25 percent overhead for the worst case benchmark. Most importantly, being a query language, SociaLite enables many more users who are not proficient in software engineering to perform network analysis easily and efficiently.

Description: In applications like social networking services and online games, multiple moving users which form a group may wish to be continuously notified about the best meeting point from their locations. A promising technique for reducing the communication frequency of the application server is to employ safe regions, which capture the validity of query results with respect to the users’ locations. Unfortunately, the safe regions in our problem exhibit characteristics such as irregular shapes and inter-dependencies, which render existing methods that compute a single safe region inapplicable to our problem. To tackle these challenges, we first examine the shapes of safe regions in our problem’s context and propose feasible approximations for them. We design efficient algorithms for computing these safe regions. We also study a variant of the problem called the sum-optimal meeting point and extend our solutions to solve this variant. Experiments with both real and synthetic data demonstrate the effectiveness of our proposal in terms of computational and communication costs.

Description: Subgraph similarity search is used in graph databases to retrieve graphs whose subgraphs are similar to a given query graph. It has been proven successful in a wide range of applications including bioinformatics and chem-informatics, etc. Due to the cost of providing efficient similarity search services on ever-increasing graph data, database outsourcing is apparently an appealing solution to database owners. Unfortunately, query service providers may be untrusted or compromised by attacks. To our knowledge, no studies have been carried out on the authentication of the search. In this paper, we propose authentication techniques that follow the popular filtering-and-verification framework. We propose an authentication-friendly metric index called ${tt GMTree}$ . Specifically, we transform the similarity search into a search in a graph metric space and derive small verification objects ( $cal VO$ s) to-be-transmitted to query clients. To further optimize ${tt GMTree}$ , we propose a sampling-based pivot selection method and an authenticated version of ${tt MCS}$ computation. Our comprehensive experiments verified the effectiveness and efficiency of our proposed techniques.

Description: The papers in this special section were presented a the 29th International Conference on Data Engineering was held in Brisbane, QLD, Australia, on April 8-11, 2013.

Description: In this paper, we study a challenging problem of deriving a taxonomy from a set of keyword phrases. A solution can benefit many real-world applications because i) keywords give users the flexibility and ease to characterize a specific domain; and ii) in many applications, such as online advertisements, the domain of interest is already represented by a set of keywords. However, it is impossible to create a taxonomy out of a keyword set itself. We argue that additional knowledge and context are needed. To this end, we first use a general-purpose knowledgebase and keyword search to supply the required knowledge and context. Then, we develop a Bayesian approach to build a hierarchical taxonomy for a given set of keywords. We reduce the complexity of previous hierarchical clustering approaches from $O(n^2; log; n)$ to $O(n; log; n)$ using a nearest-neighbor-based approximation, so that we can derive a domain-specific taxonomy from one million keyword phrases in less than an hour. Finally, we conduct comprehensive large scale experiments to show the effectiveness and efficiency of our approach. A real life example of building an insurance-related web search query taxonomy illustrates the usefulness of our approach for specific domains.

Description: In many popular applications like peer-to-peer systems, large amounts of data are distributed among multiple sources. Analysis of this data and identifying clusters is challenging due to processing, storage, and transmission costs. In this paper, we propose GDCluster, a general fully decentralized clustering method, which is capable of clustering dynamic and distributed data sets. Nodes continuously cooperate through decentralized gossip-based communication to maintain summarized views of the data set. We customize GDCluster for execution of the partition-based and density-based clustering methods on the summarized views, and also offer enhancements to the basic algorithm. Coping with dynamic data is made possible by gradually adapting the clustering model. Our experimental evaluations show that GDCluster can discover the clusters efficiently with scalable transmission cost, and also expose its supremacy in comparison to the popular method LSP2P.

Description: As new data and updates are constantly arriving, the results of data mining applications become stale and obsolete over time. Incremental processing is a promising approach to refreshing mining results. It utilizes previously saved states to avoid the expense of re-computation from scratch. In this paper, we propose i $^2$ MapReduce, a novel incremental processing extension to MapReduce, the most widely used framework for mining big data. Compared with the state-of-the-art work on Incoop, i $^2$ MapReduce (i) performs key-value pair level incremental processing rather than task level re-computation, (ii) supports not only one-step computation but also more sophisticated iterative computation, which is widely used in data mining applications, and (iii) incorporates a set of novel techniques to reduce I/O overhead for accessing preserved fine-grain computation states. We evaluate i $^2$ MapReduce using a one-step algorithm and four iterative algorithms with diverse computation characteristics. Experimental results on Amazon EC2 show significant performance improvements of i $^2$ MapReduce compared to both plain and iterative MapReduce performing re-computation.

Description: Increasing volumes of data are being produced and exchanged over the Web, in particular in tree-structured formats such as XML or JSON. This leads to a need of highly scalable algorithms and tools for processing such data, capable to take advantage of massively parallel processing platforms. This work considers the problem of efficiently parallelizing the execution of complex nested data processing, expressed in XQuery. We provide novel algorithms showing how to translate such queries into PACT, a recent framework generalizing MapReduce in particular by supporting many-input tasks. We present the first formal translation of complex XQuery algebraic expressions into PACT plans, and demonstrate experimentally the efficiency and scalability of our approach.

Description: When a “following” link is formed in a social network, will the link trigger the formation of other neighboring links? We study the diffusion phenomenon of the formation of “following” links by proposing a model to describe this link diffusion process. To estimate the diffusion strength between different links, we first conduct an analysis on the diffusion effect in 24 triadic structures and find evident patterns that facilitate the effect. We then learn the diffusion strength in different triadic structures by maximizing an objective function based on the proposed model. The learned diffusion strength is evaluated through the task of link prediction and utilized to improve the applications of follower maximization and followee recommendation, which are specific instances of influence maximization. Our experimental results reveal that incorporating diffusion patterns can indeed lead to statistically significant improvements over the performance of several alternative methods, which demonstrates the effect of the discovered patterns and diffusion model.

Description: Recent entity resolution approaches exhibit benefits when addressing the problem through unmerged duplicates : instances describing real-world objects are not merged based on apriori thresholds or human intervention, instead relevant resolution information is employed for evaluating resolution decisions during query processing using “possible worlds” semantics. In this paper, we present the first known approach for efficiently handling complex analytical queries over probabilistic databases with unmerged duplicates. We propose the entity-join operator that allows expressing complex aggregation and iceberg/top-k queries over joins between tables with unmerged duplicates and other database tables. Our technical content includes a novel indexing structure for efficient access to the entity resolution information and novel techniques for the efficient evaluation of complex probabilistic queries that retrieve analytical and summarized information over a (potentially, huge) collection of possible resolution worlds. Our extensive experimental evaluation verifies the benefits of our approach.

Description: With the increasing usage of XML on information sharing over the Internet, a mechanism for defining and enforcing XML access control is demanded, such that only authorized entities can access the sets of XML data that they are allowed to. The research interests in these areas have grown significantly in recent years. Various access control enforcement solutions have been proposed, each with its inherent advantages and disadvantages. Yet, there is still no solution that can provide superior performance in all situations. In this paper, we present HyXAC, a hybrid approach to enforce XML access control. HyXAC integrates the two most popular categories of XML access control enforcement mechanisms, and earns the benefits from both. In particular, HyXAC first preprocesses user queries by rewriting queries and removing parts violating access control rules, and evaluates the re-written queries using sub-views, if they are available. In HyXAC, views are not defined on a per-role basis. Instead, a sub-view is defined for each access control rule, and roles sharing identical rules will share sub-views. Moreover, HyXAC dynamically allocates memory and secondary storage resources to materialize and cache sub-views to improve query performance. We have conducted extensive experiments, and the results show that HyXAC improves query processing efficiency while optimizes the use of system resources.

Description: Collections of real-world data usually have implicit or explicit structural relations. For example, databases link records through foreign keys, and XML documents express associations between different values through syntax. Privacy preservation, until now, has focused either on data with a very simple structure, e.g. relational tables, or on data with very complex structure e.g. social network graphs, but has ignored intermediate cases, which are the most frequent in practice. In this work, we focus on tree structured data. Such data stem from various applications, even when the structure is not directly reflected in the syntax, e.g. XML documents. A characteristic case is a database where information about a single person is scattered amongst different tables that are associated through foreign keys. The paper defines $k^{(m,n)}$ -anonymity, which provides protection against identity disclosure and proposes a greedy anonymization heuristic that is able to sanitize large datasets. The algorithm and the quality of the anonymization are evaluated experimentally.

Description: Bag-of-words (BOW) is now the most popular way to model text in statistical machine learning approaches in sentiment analysis. However, the performance of BOW sometimes remains limited due to some fundamental deficiencies in handling the polarity shift problem. We propose a model called dual sentiment analysis (DSA), to address this problem for sentiment classification. We first propose a novel data expansion technique by creating a sentiment-reversed review for each training and test review. On this basis, we propose a dual training algorithm to make use of original and reversed training reviews in pairs for learning a sentiment classifier, and a dual prediction algorithm to classify the test reviews by considering two sides of one review. We also extend the DSA framework from polarity (positive-negative) classification to 3-class (positive-negative-neutral) classification, by taking the neutral reviews into consideration. Finally, we develop a corpus-based method to construct a pseudo-antonym dictionary, which removes DSA's dependency on an external antonym dictionary for review reversion. We conduct a wide range of experiments including two tasks, nine datasets, two antonym dictionaries, three classification algorithms, and two types of features. The results demonstrate the effectiveness of DSA in supervised sentiment classification.

Description: In the recent decades, we have witnessed the rapidly growing popularity of location-based systems. Three types of location-based queries on road networks, single-pair shortest path query, $k$ nearest neighbor ( $k$ NN) query, and keyword-based $k$ NN query, are widely used in location-based systems. Inspired by $tt R$ - $tt tree$ , we propose a height-balanced and scalable index, namely $tt G$ - $tt tree$ , to efficiently support these queries. The space complexity of $tt G$ - $tt tree$ is $mathcal {O}(|mathcal {V}|log {|mathcal {V}|})$ where ${|mathcal {V}|}$ is the number of vertices in the road network. Unlike previous works that support these queries separately, $tt G$ - $tt tree$ supports all these queries within one framework. The basis for this framework is an assembly-based method to calculate the shortest-path distances between two vertices. Based on the assembly-based method, efficient search algorithms to answer $k$ NN queries and keyword-based $k$ NN queries are developed. Experiment results show $tt G$ - $tt tree$ ’s theoretical and practical superiority over existing methods.