ALBERT

Description: Web tracking is a commonly-used practice on the Internet devoted to retrieve user information for activities such as personalization or advertisement. These techniques are said to drive the web economy, although they are commonly used to invade users' privacy. In the last years, a general concern raised about web tracking, looking forward to combat it in many ways like regulations, anti-tracking methods and even standardization. In this paper, we analyze and discuss the current techniques for web-tracking as well as techniques for its detection and analysis, and countermeasures to prevent web tracking.

Description: Smart spaces are populated by users having evolving preferences that, directly or indirectly, reflect their spatial, temporal, financial and interaction patterns of service usage. These data, if disclosed, could draw a detailed picture of users’ life in public spaces. Protection of privacy-sensitive data is especially significant in scenarios employing negotiations where many non-trusted service providers, service consumers and payment processes are involved. For such scenarios an architecture and a protocol for secure and privacy-preserving smart space usage have been developed. The presented approach relies on a trusted party operating as a public service in the ‘security infrastructure as a service’ model. The solution is designed to minimize the risk of users privacy violation from the side of service providers and attackers impersonating regular users, as well as the risk of violating privacy of users’ payment patterns from the side of payment authorities. All parties benefit from fast and secure micropayments allowing for pay-per-use model implementation, which fulfils the non-invasiveness requirement of ubiquitous services. The presented use case scenario illustrates the possible application of this approach, and adversary model explains its privacy attributes.

Description: In this article we address the problem of automated Hypertext Transfer Protocol (HTTP) request structure analysis applied to web layer cyber attacks detection. In this method, we propose a multiple HTTP sequences clustering algorithm combined with the machine-learnt classifier. The main goal behind this approach is the fact that we use the request structure and the statistical measurements of its content in order to detect anomalous behaviour of connections established between client and server. One of the advantages of the proposed method is that our solution does not need any prior knowledge about protocols and APIs that use HTTP as a transportation layer (e.g. RESTFull API, SOAP, etc.). Our experiments prove that the proposed method can achieve satisfactory results and is competitive to other state-of-the-art solutions.

Description: In recent years, mobile devices such as smartphones, tablets and wearables have become the new paradigm of user–computer interaction. The increasing use and adoption of such devices is also leading to an increased number of potential security risks. The spread of mobile malware, particularly on popular and open platforms such as Android, has become a major concern. This paper focuses on the bad-intentioned Android apps by addressing the problem of selecting the key features of such software that support the characterization of such malware. The accurate detection and characterization of this software is still an open challenge, mainly due to its ever-changing nature and the open distribution channels of Android apps. Maximum relevance minimum redundancy and evolutionary algorithms guided by information correlation measures have been applied for feature selection on the well-known Android Malware Genome (Malgenome) dataset, attaining interesting results on the most informative features for the characterization of representative families of existing Android malware.

Description: Identifying file type of file fragments has been investigated for a long time but it is still a challenge. It is found in the literature that high-entropy file fragments make the problem more complicated. Especially, existing popular file types share same compression algorithms such as deflate algorithm that causes file type identification for file fragment become harder. Applying machine learning or empirical techniques is to deal with this problem. Compression algorithms are used to reduce the size of files that have big data size and include image files. Many research work of file type identification have been done for JPEG format, and the Rate of Change feature is proven to work effectively for it. Conversely, few efforts have been made for PNG although this is a popular image format and widely used nowadays. In this article, we propose a new approach based on the deflate-encoded data detection, entropy-based clustering, and decision tree techniques to identify PNG data fragments which are the deflate-encoded fragments. Experiments showed high accuracy rates for the proposed method.

Description: Smart cards are well-known tamper-resistant devices, and due to their portability and security they represent an excellent platform for implementing strong authentication services. This contribution analyses three Spanish smart card deployments intended to be used for eGoverment services, comparing their respective contents and capabilities. As a result of our analysis, we conclude that the current deployment of smart cards for the electronic prescription service being rolled out throughout Spain could be replaced by the use of the already consolidated Spanish National Identity card.

Description: Millions of users per day are affected by unsolicited e-mail campaigns. Spam filters are capable of detecting and avoiding an increasing number of messages, but researchers have quantified a response rate of a 0.006% [ 1 ], still significant to turn a considerable profit sending millions of emails, as the spammers do. While research directions are addressing topics such as better spam filters, or spam detection inside online social networks (OSNs), in this article, we demonstrate that a classic spam model using OSN information can harvest a 7.62% of click-through rate. We collect email addresses from the Internet, complete email owner information using their public social network profile data, and analyse response of personalized spam sent to users according to their profile using a fake website. Finally, we demonstrate the effectiveness of these profile-based emails to circumvent spam detection and we compare results between typical spam and personalized spam.

Description: Much effort has been devoted to research on intrusion detection (ID) in recent years because intrusion strategies and technologies are constantly and quickly evolving. As an innovative solution based on visualization, MObile VIsualisation Connectionist Agent-Based IDS was previously proposed, conceived as a hybrid-intelligent ID System. It was designed to analyse continuous network data at a packet level and is extended in present paper for the analysis of flow-based traffic data. By incorporating clustering techniques to the original proposal, network flows are investigated trying to identify different types of attacks. The analysed real-life data (the well-known dataset from the University of Twente) come from a honeypot directly connected to the Internet (thus ensuring attack-exposure) and is analysed by means of clustering and neural techniques, individually and in conjunction. Promising results are obtained, proving the validity of the proposed extension for the analysis of network flow data.

Description: Nature inspired Swarm algorithms have proven to be effective in solving recent complex optimization problems. Comparing such algorithm is a difficult task due to many facts, the nature of the swarm, the nature of the optimization problem itself and number of controlling parameters of the swarm algorithm. In this work we compared two recent swarm algorithms applied to the community detection problem which are the Bat Algorithm (BA) and Artificial Fish Swarm Algorithm (AFSA). Community detection is an active problem in social network analysis. The problem of detecting communities can be represented as an optimization problem where a quality fitness function that captures the intuition of a community as a group of nodes with better internal connectivity than external connectivity is chosen to be optimized. We also investigated the application of the BA and AFSA in solving the community section problem. And introduced a comparative analysis between the two algorithms and other well-known methods. The study show the effectiveness and the limitations of both algorithms.

Description: Increasing attention has been paid to the prediction of earthquakes with data mining techniques during the last decade. Several works have already proposed the use of certain features serving as inputs for supervised classifiers. However, they have been successfully used without any further transformation so far. In this work, the use of principal component analysis (PCA) to reduce data dimensionality and generate new datasets is proposed. In particular, this step is inserted in a successfully already used methodology to predict earthquakes. Tokyo, one of the cities mostly threatened by large earthquakes occurrence in Japan, is studied. Several well-known classifiers combined with PCA have been used. Noticeable improvement in the results is reported.

Description: The combined application of several soft-computing and statistical techniques is proposed for the characterization of atmospheric conditions in two European regions: Madrid (Spain) and Prague (Czech Republic). The resulting Hybrid Artificial Intelligence System (HAIS) combines projection models for dimensionality reduction and clustering, combining neural and fuzzy paradigms, in a decision support tool. In present article, this proposed HAIS is applied to analyse the air quality in these two geographical regions and get a better understanding of its circumstances and evolution. To do so, real-life data from six data-acquisition stations are analysed. The main pollutants recorded at these stations between 2007 and 2014, their geographical locations and seasonal changes are all studied, in a research that shows how such factors determine variations in air-borne pollutants. Furthermore, neural projections of the clustering results from data on atmospheric pollution are studied.

Description: In database theory, the term database transformation was used to refer to a unifying treatment for computable queries and updates. Recently, it was shown that non-deterministic database transformations can be captured exactly by a variant of ASMs, the so-called Database Abstract State Machines (DB-ASMs). In this article we present a logic for DB-ASMs, extending the logic of Nanchen and Stärk for ASMs. In particular, we develop a rigorous proof system for the logic for DB-ASMs, which is proven to be sound and complete. The most difficult challenge to be handled by the extension is a proper formalization capturing non-determinism of database transformations and all its related features such as consistency, update sets or multisets associated with DB-ASM rules. As the database part of a state of database transformations is a finite structure and DB-ASMs are restricted by allowing quantifiers only over the database part of a state, we resolve this problem by taking update sets explicitly into the logic, i.e. by using an additional modal operator $[X]$, where $X$ is interpreted as an update set $Delta$ generated by a DB-ASM rule. The DB-ASM logic provides a powerful verification tool to study properties of database transformations.

Description: Data Clustering aims to discover groups within the data based on similarities, with a minimal, if any, knowledge of their structure. Variations in the results may occur due to many factors, including algorithm parameters, initialization and stopping criteria. The usage of different attributes or even different subsets of data usually lead to different results. Self-organizing maps (SOM) has been widely used for a variety of tasks regarding data analysis, including data visualization and clustering. A machine committee, or ensemble, is a set of neural networks working independently with some system that enable the combination of individual results into a single output, with the aim to achieve a better generalization compared to a unique neural network. This article presents a new ensemble method that uses SOM networks. Cluster validity indexes are used to combine neuron weights from different maps with different sizes. Results are shown from simulations with real and synthetic data, from the UCI Repository and Fundamental Clustering Problems Suite. The proposed method presented promising results, with increased performance compared with conventional single Kohonen map.

Description: This work addresses the problem of energy-efficient scheduling and allocation of tasks in multicore environments, where the tasks can allow a certain loss in accuracy in the output, while still providing proper functionality and meeting an energy budget. This margin for accuracy loss is exploited by using computing techniques that reduce the work load, and thus can also result in significant energy savings. To this end, we use the technique of loop perforation, that transforms loops to execute only a subset of their original iterations, and integrate this technique into our existing optimization tool for energy-efficient scheduling. To verify that a schedule meets an energy budget, both safe upper and lower bounds on the energy consumption of the tasks involved are needed. For this reason, we use a parametric approach to estimate safe (and tight) energy bounds that are practical for energy verification (and optimization applications). This approach consists in dividing a program into basic (‘branchless’) blocks, establishing the maximal (resp. minimal) energy consumption for each block using an evolutionary algorithm, and combining the obtained values according to the program control flow, by using static analysis to produce energy bound functions on input data sizes. The scheduling tool uses evolutionary algorithms coupled with the energy bound functions for estimating the energy consumption of different schedules. The experiments with our prototype implementation were performed on multicore XMOS chips, but our approach can be adapted to any multicore environment with minor changes. The experimental results show that our new scheduler enhanced with loop perforation improves on the previous one, achieving significant energy savings (31% on average for the test programs) for acceptable levels of accuracy loss.

Description: The finite model property is a key step in proving decidability of modal logics. By adapting the filtration method to the generalized Veltman semantics for interpretability logics, we have been able to prove the finite model property of interpretability logic ILM0 w.r.t. generalized Veltman models. We use the same technique to prove the finite model property of interpretability logic ILW* w.r.t. generalized Veltman models. The missing link needed to prove the decidability of ILM0 was completeness w.r.t. generalized Veltman models, which we obtain in this article. Thus, we prove the decidability of ILM0, which was an open problem. Using the same technique, we prove that ILW* is also decidable.

Description: Facility location decisions are critical in strategic planning for a wide range of operational and logistical decisions. Facility location problem with focus on logistics distribution centre (LDC) in Balkan Peninsula (BP) is discussed in this article. Methodological hybrid genetic algorithm, Analytical Hierarchy Process, and fuzzy c-means method is proposed here and it is shown how such a model can be of assistance in analysing a multi criteria decision-making problem. This research represents continuation of three existing studies. The experimental results in our research could be well compared with other official results of the feasibility study of the LDC located in BP.

Description: Although logical consistency is desirable in scientific research, standard statistical hypothesis tests are typically logically inconsistent. To address this issue, previous work introduced agnostic hypothesis tests and proved that they can be logically consistent while retaining statistical optimality properties. This article characterizes the credal modalities in agnostic hypothesis tests and uses the hexagon of oppositions to explain the logical relations between these modalities. Geometric solids that are composed of hexagons of oppositions illustrate the conditions for these modalities to be logically consistent. Prisms composed of hexagons of oppositions show how the credal modalities obtained from two agnostic tests vary according to their threshold values. Nested hexagons of oppositions summarize logical relations between the credal modalities in these tests and prove new relations.

Description: This article presents several methods used in motion capture to measure jumps. The traditional systems to acquire jump information are force plates, but they are very expensive to most people. Amateur sports enthusiasts who want to improve their performance, do not have enough money to spend in professional systems ($+/-20.000$EUR). The price reduction of electronic devices, specifically the inertial measurement units (IMU), are generating new methods of motion capture. In this article we present the state-of the-art motion capture systems for this purpose, from the classical force plates to latest released IMUs. Experiments show that the IMU is equally valid for measuring vertical jump.

Description: This article presents an Off-line handwritten digit recognition approach based on neural networks. We define a numeric character as a composition of vertical and horizontal strokes. After the preprocessing, we use dynamic zoning to retrieve the positions where vertical strokes – the main strokes — are joined to horizontal strokes. These features are recorded into a representative string and verified using a custom matching pattern. Finally, a multilayer perceptron neural network is fed with the previous data to raise the learning process. The results gathered from the experiments performed on the well-known MNIST handwritten database are compared against other proposals providing promising results.

Description: This article introduces a new formal theory, the theory of $M$-system, which represents a many-valued logic system. Basic terms and concepts are defined which are the foundation for their further application. Though the theory comes from the field of electric circuitry, an attempt will be made to extend it to applications in linguistics. To check the examples used in this article, a Haskell program was made, which can be found at https://www.schoolofhaskell.com/user/ivazler/linguistic.

Description: Nowadays, it is common for higher education institutions to use computer-based exams, partly or integrally, in their evaluation processes. These exams, much like their paper-based counterparts, are one of the most significant sources of stress in the life of students. However, the fact that exams are undertaken in a computer allows for new features to be acquired that may provide more reliable insights into the behaviour and state of the student during the exam. In this article we analyse these novel behavioural features and explore, to which extent, they can point out previously unknown phenomena. Specifically, we show that the time a student takes to complete an exam is correlated with mouse dynamics features. In practical terms, we are able to predict the duration of each individual exam with a satisfying error based on the interaction patterns of the student.