ALBERT

Description: Undesirable vibrations resulting from the use of vibrating hand-held tools decrease the tool performance and user productivity. In addition, prolonged exposure to the vibration can cause ergonomic injuries known as the hand-arm vibration syndrome (HVAS). Therefore, it is very important to design a vibration suppression mechanism that can isolate or suppress the vibration transmission to the users’ hands to protect them from HAVS. While viscoelastic materials in anti-vibration gloves are used as the passive control approach, an active vibration control has shown to be more effective but requires the use of sensors, actuators and controllers. In this paper, the design of a controller for an anti-vibration glove is presented. The aim is to keep the level of vibrations transferred from the tool to the hands within a healthy zone. The paper also describes the formulation of the hand-glove system’s mathematical model and the design of a fuzzy parallel distributed compensation (PDC) controller that can cater for different hand masses. The performances of the proposed controller are evaluated through simulations and the results are benchmarked with two other active vibration control techniques-proportional integral derivative (PID) controller and active force controller (AFC). The simulation results show a superior performance of the proposed controller over the benchmark controllers. The designed PDC controller is able to suppress the vibration transferred to the user’s hand 93% and 85% better than the PID controller and the AFC, respectively.

Description: Fault prediction is a necessity to deliver high-quality software. The absence of training data and mechanism to labeling a cluster faulty or fault-free is a topic of concern in software fault prediction (SFP). Inheritance is an important feature of object-oriented development, and its metrics measure the complexity, depth, and breadth of software. In this paper, we aim to experimentally validate how much inheritance metrics are helpful to classify unlabeled data sets besides conceiving a novel mechanism to label a cluster as faulty or fault-free. We have collected ten public data sets that have inheritance and C&K metrics. Then, these base datasets are further split into two datasets labeled as C&K with inheritance and the C&K dataset for evaluation. K-means clustering is applied, Euclidean formula to compute distances and then label clusters through the average mechanism. Finally, TPR, Recall, Precision, F1 measures, and ROC are computed to measure performance which showed an adequate impact of inheritance metrics in SFP specifically classifying unlabeled datasets and correct classification of instances. The experiment also reveals that the average mechanism is suitable to label clusters in SFP. The quality assurance practitioners can benefit from the utilization of metrics associated with inheritance for labeling datasets and clusters.

Description: Multitask learning has led to significant advances in Natural Language Processing, including the decaNLP benchmark where question answering is used to frame 10 natural language understanding tasks in a single model. In this work we show how models trained to solve decaNLP fail with simple paraphrasing of the question. We contribute a crowd-sourced corpus of paraphrased questions (PQ-decaNLP), annotated with paraphrase phenomena. This enables analysis of how transformations such as swapping the class labels and changing the sentence modality lead to a large performance degradation. Training both MQAN and the newer T5 model using PQ-decaNLP improves their robustness and for some tasks improves the performance on the original questions, demonstrating the benefits of a model which is more robust to paraphrasing. Additionally, we explore how paraphrasing knowledge is transferred between tasks, with the aim of exploiting the multitask property to improve the robustness of the models. We explore the addition of paraphrase detection and paraphrase generation tasks, and find that while both models are able to learn these new tasks, knowledge about paraphrasing does not transfer to other decaNLP tasks.

Description: Over the last few years, private and public organizations have suffered an increasing number of cyber-attacks owing to excessive exploitation of technological vulnerabilities. The major objective of these attacks is to gain illegal profits by extorting organizations which adversely impact their normal operations and reputation. To mitigate the proliferation of attacks, it is significant for manufacturers to evaluate their IT products through a set of security-related functional and assurance requirements. Common Criteria (CC) is a well-recognized international standard, focusing on ensuring security functionalities of an IT product along with the special emphasis on IS design and life-cycle. Apart from this, it provides a list of assurance classes, families, component, and elements based on which security EALs can be assigned to IT products. In this survey, we have provided a quick overview of the CC followed by the analysis of country-specific implementation of CC schemes to develop an understanding of critical factors. These factors play a significant role by providing assistance in IT products evaluation in accordance with CC. To serve this purpose, a comprehensive comparative analysis of four schemes belonging to countries including US, UK, Netherlands, and Singapore has been conducted. This comparison has aided to propose best practices for realizing an efficient and new CC scheme for the countries which have not designed it yet and for improving the existing CC schemes. Finally, we conclude the paper by providing some future directions regarding automation of the CC evaluation process.

Description: The proposed research motivates the 6G cellular networking for the Internet of Everything’s (IoE) usage empowerment that is currently not compatible with 5G. For 6G, more innovative technological resources are required to be handled by Mobile Edge Computing (MEC). Although the demand for change in service from different sectors, the increase in IoE, the limitation of available computing resources of MEC, and intelligent resource solutions are getting much more significant. This research used IScaler, an effective model for intelligent service placement solutions and resource scaling. IScaler is considered to be made for MEC in Deep Reinforcement Learning (DRL). The paper has considered several requirements for making service placement decisions. The research also highlights several challenges geared by architectonics that submerge an Intelligent Scaling and Placement module.

Description: This paper continues the work initiated by the authors on the feasibility of using ParaView as visualization software for the analysis of parallel Computational Fluid Dynamics (CFD) codes’ performance. Current performance tools have limited capacity of displaying their data on top of three-dimensional, framed (i.e., time-stepped) representations of the cluster’s topology. In our first paper, a plugin for the open-source performance tool Score-P was introduced, which intercepts an arbitrary number of manually selected code regions (mostly functions) and send their respective measurements–amount of executions and cumulative time spent–to ParaView (through its in situ library, Catalyst), as if they were any other flow-related variable. Our second paper added to such plugin the capacity to (also) map communication data (messages exchanged between MPI ranks) to the simulation’s geometry. So far the tool was limited to codes which already have the in situ adapter; but in this paper, we will take the performance data and display it–also in codes without in situ–on a three-dimensional representation of the hardware resources being used by the simulation. Testing is done with the Multi-Grid and Block Tri-diagonal NPBs, as well as Rolls-Royce’s CFD code, Hydra. The benefits and overhead of the plugin's new functionalities are discussed.

Description: The presence of abusive and vulgar language in social media has become an issue of increasing concern in recent years. However, research pertaining to the prevalence and identification of vulgar language has remained largely unexplored in low-resource languages such as Bengali. In this paper, we provide the first comprehensive analysis on the presence of vulgarity in Bengali social media content. We develop two benchmark corpora consisting of 7,245 reviews collected from YouTube and manually annotate them into vulgar and non-vulgar categories. The manual annotation reveals the ubiquity of vulgar and swear words in Bengali social media content (i.e., in two corpora), ranging from 20% to 34%. To automatically identify vulgarity, we employ various approaches, such as classical machine learning (CML) classifiers, Stochastic Gradient Descent (SGD) optimizer, a deep learning (DL) based architecture, and lexicon-based methods. Although small in size, we find that the swear/vulgar lexicon is effective at identifying the vulgar language due to the high presence of some swear terms in Bengali social media. We observe that the performances of machine leanings (ML) classifiers are affected by the class distribution of the dataset. The DL-based BiLSTM (Bidirectional Long Short Term Memory) model yields the highest recall scores for identifying vulgarity in both datasets (i.e., in both original and class-balanced settings). Besides, the analysis reveals that vulgarity is highly correlated with negative sentiment in social media comments.

Description: Anomaly detection in computer networks is a complex task that requires the distinction of normality and anomaly. Network attack detection in information systems is a constant challenge in computer security research, as information systems provide essential services for enterprises and individuals. The consequences of these attacks could be the access, disclosure, or modification of information, as well as denial of computer services and resources. Intrusion Detection Systems (IDS) are developed as solutions to detect anomalous behavior, such as denial of service, and backdoors. The proposed model was inspired by the behavior of dendritic cells and their interactions with the human immune system, known as Dendritic Cell Algorithm (DCA), and combines the use of Multiresolution Analysis (MRA) Maximal Overlap Discrete Wavelet Transform (MODWT), as well as the segmented deterministic DCA approach (S-dDCA). The proposed approach is a binary classifier that aims to analyze a time-frequency representation of time-series data obtained from high-level network features, in order to classify data as normal or anomalous. The MODWT was used to extract the approximations of two input signal categories at different levels of decomposition, and are used as processing elements for the multi resolution DCA. The model was evaluated using the NSL-KDD, UNSW-NB15, CIC-IDS2017 and CSE-CIC-IDS2018 datasets, containing contemporary network traffic and attacks. The proposed MRA S-dDCA model achieved an accuracy of 97.37%, 99.97%, 99.56%, and 99.75% for the tested datasets, respectively. Comparisons with the DCA and state-of-the-art approaches for network anomaly detection are presented. The proposed approach was able to surpass state-of-the-art approaches with UNSW-NB15 and CSECIC-IDS2018 datasets, whereas the results obtained with the NSL-KDD and CIC-IDS2017 datasets are competitive with machine learning approaches.

Description: Epilepsy is a common neurological disease that affects a wide range of the world population and is not limited by age. Moreover, seizures can occur anytime and anywhere because of the sudden abnormal discharge of brain neurons, leading to malfunction. The seizures of approximately 30% of epilepsy patients cannot be treated with medicines or surgery; hence these patients would benefit from a seizure prediction system to live normal lives. Thus, a system that can predict a seizure before its onset could improve not only these patients’ social lives but also their safety. Numerous seizure prediction methods have already been proposed, but the performance measures of these methods are still inadequate for a complete prediction system. Here, a seizure prediction system is proposed by exploring the advantages of multivariate entropy, which can reflect the complexity of multivariate time series over multiple scales (frequencies), called multivariate multiscale modified-distribution entropy (MM-mDistEn), with an artificial neural network (ANN). The phase-space reconstruction and estimation of the probability density between vectors provide hidden complex information. The multivariate time series property of MM-mDistEn provides more understandable information within the multichannel data and makes it possible to predict of epilepsy. Moreover, the proposed method was tested with two different analyses: simulation data analysis proves that the proposed method has strong consistency over the different parameter selections, and the results from experimental data analysis showed that the proposed entropy combined with an ANN obtains performance measures of 98.66% accuracy, 91.82% sensitivity, 99.11% specificity, and 0.84 area under the curve (AUC) value. In addition, the seizure alarm system was applied as a postprocessing step for prediction purposes, and a false alarm rate of 0.014 per hour and an average prediction time of 26.73 min before seizure onset were achieved by the proposed method. Thus, the proposed entropy as a feature extraction method combined with an ANN can predict the ictal state of epilepsy, and the results show great potential for all epilepsy patients.

Description: Background The planning and control of wind power production rely heavily on short-term wind speed forecasting. Due to the non-linearity and non-stationarity of wind, it is difficult to carry out accurate modeling and prediction through traditional wind speed forecasting models. Methods In the paper, we combine empirical mode decomposition (EMD), feature selection (FS), support vector regression (SVR) and cross-validated lasso (LassoCV) to develop a new wind speed forecasting model, aiming to improve the prediction performance of wind speed. EMD is used to extract the intrinsic mode functions (IMFs) from the original wind speed time series to eliminate the non-stationarity in the time series. FS and SVR are combined to predict the high-frequency IMF obtained by EMD. LassoCV is used to complete the prediction of low-frequency IMF and trend. Results Data collected from two wind stations in Michigan, USA are adopted to test the proposed combined model. Experimental results show that in multi-step wind speed forecasting, compared with the classic individual and traditional EMD-based combined models, the proposed model has better prediction performance. Conclusions Through the proposed combined model, the wind speed forecast can be effectively improved.