Journal Cover
Information Sciences
Journal Prestige (SJR): 1.635
Citation Impact (citeScore): 5
Number of Followers: 614  
 
  Hybrid Journal Hybrid journal (It can contain Open Access articles)
ISSN (Print) 0020-0255
Published by Elsevier Homepage  [3182 journals]
  • A Smart Fault-detection Approach with Feature Production and Extraction
           Processes
    • Abstract: Publication date: Available online 12 November 2019Source: Information SciencesAuthor(s): Shih-Yu Li, Kai-Ren Gu In this paper, a smart fault-detection approach with feature production and extraction procedures is developed for industrial ball bearing systems. By developing a dynamic nonlinear error system composed of a main system and data-feeding system with appropriate parameters, the vibration signals of different fault states captured from ball bearings in the time domain can be mathematically mapped to the chaotic domain for feature production. Furthermore, through the designed feature extraction process, the relevant Euclidean feature values (EFVs) can be obtained for the classification of four different fault states. Three fault conditions at diameters of 7 mil, 14 mil, and 21 mil and a depth of 0.011 inches are illustrated for performance investigations. The experimental results show that the proposed smart detection approach is effective and feasible for identifying different fault states in real time.
       
  • NoteSum: An Integrated Note Summarization System by Using Text Mining
           Algorithms
    • Abstract: Publication date: Available online 12 November 2019Source: Information SciencesAuthor(s): Hei-Chia Wang, Wei-Fan Chen, Chen-Yu Lin This study implemented an integrated system of Note Summarization (NoteSum) that merged with multi-users’ notes and searched for relevant information on the Internet and, slides, and textbooks to create a summary for students to learn effectively. The integrated system's framework consists of four different modules: Topic Identification Module, Supporting Material Finding Module, Content Mapping Module and Learning Material Integrating Module. Five experiments were conducted; these resulted in the following findings. First, translating notes with the assistance of topic terms could enhance translation quality. Second, when mapping contents, NoteSum performed better in a discussion-based course rather than in a technical course. Third, the Jensen-Shannon (JS) Divergence was used to assess the generated summary that performed better for the discussion-based course. Fourth, the three attributes—presence of topic terms, number of non-topic words, and ratio of the words with important parts of speech—had different effects on different subjects. Finally, we compared NoteSum with other existing summarization systems. The results indicated that the NoteSum-generated summary was closer to students’ original notes and thus resulted in better performance in readability, informativeness, and completeness. All the results confirm that our proposed NoteSum is an effective note summarization system for student learning.
       
  • Data Imbalance in Classification: Experimental Evaluation
    • Abstract: Publication date: Available online 11 November 2019Source: Information SciencesAuthor(s): Fadi Thabtah, Suhel Hammoud, Firuz Kamalov, Amanda H. Gonsalvesv The advent of Big Data has ushered a new era of scientific breakthroughs. One of the common issues that affects raw data is class imbalance problem which refers to imbalanced distribution of values of the response variable. This issue is present in fraud detection, network intrusion detection, medical diagnostics, and a number of other fields where negatively labeled instances significantly outnumber positively labeled instances. Modern machine learning techniques struggle to deal with imbalanced data by focusing on minimizing the error rate for the majority class while ignoring the minority class. The goal of our paper is demonstrate the effects of class imbalance on classification models. Concretely, we study the impact of varying class imbalance ratios on classifier accuracy. By highlighting the precise nature of the relationship between the degree of class imbalance and the corresponding effects on classifier performance we hope to help researchers to better tackle the problem. To this end, we carry out extensive experiments using 10-fold cross validation on a large number of datasets. Our analysis confirms that class imbalance has a significant negative impact on the performance of a classifier. In particular, we determine that the relationship between the class imbalance ratio and the accuracy is convex. We also find that addressing the issue of class imbalance in the pre-training phase can have a substantial positive impact on the accuracy of a classifier.
       
  • Multi-modal medical image segmentation based on vector-valued active
           contour models
    • Abstract: Publication date: Available online 9 November 2019Source: Information SciencesAuthor(s): Lingling Fang, Xin Wang, Lujie Wang Positron emission tomography (PET), magnetic resonance imaging (MRI) and computed tomography (CT) are widely utilized medical imaging modalities that provide essential anatomic and structural details. Many medical segmentation methods are not effective for a single-modal image of poor quality (e.g., low contrast in CT or low spatial resolution in PET). For practical radiotherapy treatment planning, multi-modal imaging information is regularly used. In this paper, a novel vector-valued active contour model is proposed to segment multi-modal medical images simultaneously for abnormal tissue regions. The method makes use of the functionality information and anatomical structure information advantages from each modality. Since each modality has its own signal characteristics, we use region-based information, combining hybrid mean intensities simultaneously. Furthermore, by utilizing a two-dimensional vector field with different image modalities, edge-based information is used to constrain the results of the image segmentation. The proposed approach is evaluated on datasets including lung PET-CT and brain MRI-CT images. Our qualitative and quantitative research results confirm the effectiveness of the proposed method.
       
  • Fast finite-time Adaptive Stabilization of High-Order Uncertain Nonlinear
           Systems with Output Constraint and Zero Dynamics
    • Abstract: Publication date: Available online 9 November 2019Source: Information SciencesAuthor(s): Zong-Yao Sun, Cheng-Qian Zhou, Chih-Chiang Chen, Qinghua Meng This paper investigates the problem of fast finite-time adaptive stabilization for a class of high-order uncertain nonlinear systems with an output constraint and zero dynamics. A continuous stabilizer with an adaptive mechanism is constructed by utilizing a tangent function and a serial of nonnegative integral functions equipped with sign functions, which guarantees the system output to be restricted in a pre-specified region and a faster convergence speed of system states compared to traditional finite-time stabilizers. The main novelty of this paper is the skillful selection of Lyapunov functions and the new perspective of constructing a fast finite-time adaptive stabilizer with the consideration of output constraints as well as dynamic and parameter uncertainties. A simple example is given to demonstrate the effectiveness of the proposed strategy.
       
  • The data richness estimation framework for federated data warehouse
           integration
    • Abstract: Publication date: Available online 9 November 2019Source: Information SciencesAuthor(s): Rafał Kern, Adrianna Kozierkiewicz, Marcin Pietranik A federated data warehouse is a tool that provides an end-user a unified perspective on a finite set of independent data warehouses. This requires creating a global schema from partial schemas, which remains purely virtual. This is a result of iterative integration of participating data warehouses. It is then used to simulate that the aforementioned set of participating warehouses as an effectively one, “super” data warehouse exposed to the end-user. In this paper, authors present a framework, that can be used to evaluate the profitability of adding a new data warehouse to the existing federation, in terms of increased data richness and its expressiveness. Solid formal foundations are provided, along with heuristic algorithms, an experimental verification (which involved two different experimental procedures) and a statistical analysis of obtained results.
       
  • H ∞ leader-following consensus of nonlinear multi-agent systems under
           semi-Markovian switching topologies with partially unknown transition
           rates
    • Abstract: Publication date: Available online 9 November 2019Source: Information SciencesAuthor(s): Minhong He, Jingru Mu, Xiaowu Mu This paper investigates the H∞ leader-following consensus problem for nonlinear multi-agent systems under semi-Markovian switching topologies. The switching of the topologies is governed by a semi-Markovian jump process, which covers a Markovian jump process as a special case. In many practical systems, it is difficult to obtain the transition rate matrix, thus the transition rates are considered to be not completely known in the paper. External perturbations are considered in the paper, and H∞ control theory is applied. By utilising stochastic technique, sufficient conditions expressed in terms of linear matrix inequalities are derived to ensure that the H∞ leader-following consensus can be reached with a prescribed performance index. Finally, a numerical example is given to show the effectiveness of the theoretical results.
       
  • Robust and blind image watermarking in DCT domain using inter-block
           coefficient correlation
    • Abstract: Publication date: Available online 9 November 2019Source: Information SciencesAuthor(s): Hung-Jui Ko, Cheng-Ta Huang, Gwoboa Horng, Shiuh-Jeng WANG This study presents a robust and transparent watermarking method that exploits block-based discrete cosine transform (DCT) coefficient modification. The difference in the DCT coefficients of two blocks is calculated and modified based on the watermark bit to adjust this difference to a predefined range. The first coefficient in the upper left corner of the array basis function is known as the direct current (DC) coefficient, whereas the remainder includes the alternating current (AC) coefficients. The extent of the DCT coefficient modifications depends on the DC coefficient and median of the AC coefficients ordered by a zig-zag sequence. The robustness of the proposed method against various attacks was evaluated experimentally, and experimental results demonstrate that the proposed method possesses great robustness against various single and combined attacks.
       
  • Efficient Temporal Core Maintenance of Massive Graphs
    • Abstract: Publication date: Available online 9 November 2019Source: Information SciencesAuthor(s): Wen Bai, Yadi Chen, Di Wu k−core is a cohesive subgraph such that every vertex has at least k neighbors within the subgraph, which provides a good measure to evaluate the importance of vertices as well as their connections. Unfortunately, k−core cannot adequately reveal the structure of a temporal graph, in which two vertices may connect multiple edges containing time information. As a result, (k,h)−core is derived from k−core, which is also called temporal core, to provide a well-formulated definition, where h represents the number of temporal edges between two vertices. However, it is costly to repeatedly decompose a temporal graph changing over time.To address this challenge, we study the method of (k,h)−core maintenance, which can find current (k,h)−cores with less computational efforts. To estimate the influence scope of inserted (removed) edges, we propose quasi-temporal core, denoted by quasi−(k,h)−core, which relaxes the constraint of (k,h)−core but still has similar properties to (k,h)−core. With the aid of quasi−(k,h)−core, our insertion algorithm finds the minimum incremental graph for each influenced (k,h)−core, and the removal algorithm adjusts each influenced (k,h)−core in the minimal range. Experimental results verify effectiveness and scalability of our proposed algorithms.
       
  • Semantics-Aware Influence Maximization in Social Networks
    • Abstract: Publication date: Available online 9 November 2019Source: Information SciencesAuthor(s): Yipeng Chen, Qiang Qu, Yuanxiang Ying, Hongyan Li, Jialie Shen Influence Maximization (IM) plays an essential role in various social network applications. One such application is viral marketing to trigger a large cascade of product adoption from a small number of users by utilizing “Word-of-Mouth” effect in social networks. IM aims to return a set of users that can influence the largest fraction of a network, such as the early user who demonstrates the good features of a product in marketing. The traditional IM algorithms treat all users equally and ignore semantic context associated with the users, though it has been studied previously. To consider the semantics, we introduce a semantics-aware influence maximization (SIM) problem. The SIM problem integrates semantic information of users with influence maximization by measuring influence spread based on semantic values under a given model, and it aims to find a set of users that maximizes the influence spread, shown to be NP-hard. Generalized Reverse Influence Set based framework for SIM problems (GRIS-SIM) is used to solve SIM with different semantics, which provides a (1−1/e−ε)-approximation solution for each SIM instance. To our knowledge, the guarantee is state-of-the-art in the IM studies. GRIS-SIM enables auto-generation of sampling strategies for various social networks. In this study, we also present three sampling strategies that can be generated to achieve the best approximation guarantee, and one of the three is proved to be the optimal strategy by having the same performance guarantee within the optimal time. Furthermore, in order to show the generality and effectiveness of the proposed GRIS technique, we apply it into solving other IM problems (e.g., the distance-aware influence maximization, DAIM). Extensive experiments on both real-life and synthetic datasets demonstrate the effectiveness, efficiency, and scalability of our methods. The results on large real data show that GRIS-SIM is able to achieve 58% improvement on average in expected influence compared with rivals, and the method adopting GRIS can achieve 65% improvement on average.
       
  • Adaptive neural control for non-strict-feedback nonlinear systems with
           input delay
    • Abstract: Publication date: Available online 9 November 2019Source: Information SciencesAuthor(s): Huanqing Wang, Siwen Liu, Xuebo Yang In this research, the controller design problem is considered for non-strict-feedback systems with input delay. An appropriate auxiliary system is utilized to deal with the difficulties appeared in input delay. By the utilization of backstepping and adaptive neural control, a state-feedback stabilization controller is developed. The designed controller enables the variables in the closed-loop system to be semi-globally uniformly ultimately bounded in a small interval around the origin. The main significance of this research is that an intelligent control scheme is extended to a class of nonlinear systems with non-strict-feedback form and input delay, simultaneously. Finally, an example is given to show the effectiveness of the control method.
       
  • A Hybrid Deep Learning Model for Efficient Intrusion Detection in Big Data
           Environment
    • Abstract: Publication date: Available online 8 November 2019Source: Information SciencesAuthor(s): Mohammad Mehedi Hassan, Abdu Gumaei, Ahmed Alsanad, Majed Alrubaian, Giancarlo Fortino The volume of network and Internet traffic is expanding daily, with data being created at the zettabyte to petabyte scale at an exceptionally high rate. These can be characterized as big data, because they are large in volume, variety, velocity, and veracity. Security threats to networks, the Internet, websites, and organizations are growing alongside this growth in usage. Detecting intrusions in such a big data environment is difficult. Various intrusion-detection systems (IDSs) using artificial intelligence or machine learning have been proposed for different types of network attacks, but most of these systems either cannot recognize unknown attacks or cannot respond to such attacks in real time. Deep learning models, recently applied to large-scale big data analysis, have shown remarkable performance in general but have not been examined for detection of intrusions in a big data environment. This paper proposes a hybrid deep learning model to efficiently detect network intrusions based on a convolutional neural network (CNN) and a weight-dropped, long short-term memory (WDLSTM) network. We use the deep CNN to extract meaningful features from IDS big data and WDLSTM to retain long-term dependencies among extracted features to prevent overfitting on recurrent connections. The proposed hybrid method was compared with traditional approaches in terms of performance on a publicly available dataset, demonstrating its satisfactory performance.
       
  • OCam:+Out-of-core+Coordinate+Descent+Algorithm+for Matrix Completion&rft.title=Information+Sciences&rft.issn=0020-0255&rft.date=&rft.volume=">OCam: Out-of-core Coordinate Descent Algorithm for Matrix Completion
    • Abstract: Publication date: Available online 7 November 2019Source: Information SciencesAuthor(s): Dongha Lee, Jinoh Oh, Hwanjo Yu Recently, there are increasing reports that most datasets can be actually stored in disks of a single off-the-shelf workstation, and utilizing out-of-core methods is much cheaper and even faster than using a distributed system. For these reasons, out-of-core methods have been actively developed for machine learning and graph processing. The goal of this paper is to develop an efficient out-of-core matrix completion method based on coordinate descent approach. Coordinate descent-based matrix completion (CD-MC) has two strong benefits over other approaches: 1) it does not involve heavy computation such as matrix inversion and 2) it does not have step-size hyper-parameters, which reduces the effort for hyper-parameter tuning. Existing solutions for CD-MC have been developed and analyzed for in-memory setting and they do not take disk-I/O into account. Thus, we propose OCam, a novel out-of-core coordinate descent algorithm for matrix completion. Our evaluation results and cost analyses provide sound evidences supporting the following benefits of OCam: (1) Scalability – OCam is a truly scalable out-of-core method and thus decomposes a matrix larger than the size of memory, (2) Efficiency – OCam is super fast. OCam is up to 10x faster than the state-of-the-art out-of-core method, and up to 4.1x faster than a competing distributed method when using eight machines. The source code of OCam will be available for reproducibility.
       
  • Robust subspace clustering based on non-convex low-rank approximation and
           adaptive kernel
    • Abstract: Publication date: Available online 6 November 2019Source: Information SciencesAuthor(s): Xuqian Xue, Xiaoqian Zhang, Xinghua Feng, Huaijiang Sun, Wei Chen, Zhigui Liu As a relatively advanced method, the low-rank kernel space clustering method shows good performance in dealing with nonlinear structure of high-dimensional data. Unfortunately, this method is sensitive to large corruptions and doesn’t balance the contribution of all singular values. To solve the above problems, the low-rank kernel method is modified, and a robust subspace clustering method (LAKRSC) based on non-convex low-rank approximation and adaptive kernel is proposed. In our model, the weighted Schatten p-norm is introduced to balance the importance of different singular values, which can more accurately approximate the rank function and be more flexible in practical applications. Therefore, applying weighted Schatten p-norm to adaptive kernel can approximate the original low rank hypothesis better when the data is mapped into the feature space. In addition, our model uses correntropy to handle complex noise which enhances the robustness of the model. A new algorithm HQ&ADMM, combined by Half-Quadratic technique (HQ) and ADMM, is studied to solve our model. Experiments on four real-world datasets show that the clustering performance of LAKRSC is significantly better than that of several more advanced methods.
       
  • Image Representation of Pose-Transition Feature for 3D Skeleton-Based
           Action Recognition
    • Abstract: Publication date: Available online 5 November 2019Source: Information SciencesAuthor(s): Thien Huynh-The, Cam-Hao Hua, Trung-Thanh Ngo, Dong-Seong Kim Recently, skeleton-based human action recognition has received more interest from industrial and research communities for many practical applications thanks to the popularity of depth sensors. A large number of conventional approaches, which have exploited handcrafted features with traditional classifiers, cannot learn high-level spatiotemporal features to precisely recognize complex human actions. In this paper, we introduce a novel encoding technique, namely Pose-Transition Feature to Image (PoT2I), to transform skeleton information to image-based representation for deep convolutional neural networks (CNNs). The spatial joint correlations and temporal pose dynamics of an action are exhaustively depicted by an encoded color image. For learning action models, we fine-tune end-to-end a pre-trained network to thoroughly capture multiple high-level features at multi-scale action representation. The proposed method is benchmarked on several challenging 3D action recognition datasets (e.g., UTKinect-Action3D, SBU-Kinect Interaction, and NTU RGB+D) with different parameter configurations for performance analysis. Outstanding experimental results with the highest accuracy of 90.33% on the most challenging NTU RGB+D dataset demonstrate that our action recognition method with PoT2I outperforms state-of-the-art approaches.
       
  • Virtual Linguistic Trust Degree-Based Evidential Reasoning Approach and
           Its Application to Emergency Response Assessment of Railway Station
    • Abstract: Publication date: Available online 5 November 2019Source: Information SciencesAuthor(s): Jianmei Ye, Zeshui Xu, Xunjie Gou In real life, the assessment information is usually expressed by virtual linguistic terms, but the trust degrees of different assessment values to the same reference grade are usually ignored. In addition, the provided information is usually incomplete or uncertain, because the objective things’ characteristics are elusive and the decision maker's knowledge is limited. To solve such problems, in this paper, we first propose the virtual linguistic trust degree. Then, in view of the nonlinear changes of the decision maker's psychological states, we depict the virtual linguistic trust degree by a nonlinear function, and present some relevant aggregation operators. In order to avoid the loss of information, we propose a novel evidential reasoning approach that combines the virtual linguistic trust degree with basic unit-interval monotonic function. In this approach, the normalized basic probability mass can be obtained by the continuous basic unit-interval monotonic function, which satisfies the consensus axiom of original evidential reasoning. Meanwhile, we propose the normalized factor on the basis of the rule that the remaining unassigned probability mass is assigned into any subsets, and then present the whole framework of an extended evidential reasoning algorithm. Furthermore, a numerical example about the emergency response assessment of railway station is conducted to show the usage of the algorithm. Finally, the validity of this algorithm is demonstrated by the comparative analysis.
       
  • Resilient and secure remote monitoring for a class of cyber-physical
           systems against attacks
    • Abstract: Publication date: Available online 5 November 2019Source: Information SciencesAuthor(s): Xiaohua Ge, Qing-Long Han, Xian-Ming Zhang, Derui Ding, Fuwen Yang This paper is concerned with the resilient and secure remote monitoring of a cyber-physical system of a discrete time-varying state-space form against attacks. The specific statistical characteristic, magnitude, occurring place and time of the attack signals are not required during the monitor design and attack detection procedures. First, an optimal ellipsoidal state prediction and estimation method is delicately developed in such a way that the recursively computed prediction ellipsoid and estimate ellipsoid can both guarantee the containment of the true system state at each time step regardless of the unknown but bounded input signal. It is expected that the two ellipsoids can resist certain attacks as the calculated state prediction and state estimate are sets in state-space rather than single pointwise vectors, thus potentially enhancing the resilience of the remote monitoring system. Second, a set-based evaluation mechanism in combination with a remedy measure are proposed to provide timely detection of certain attacks. Furthermore, a numerically efficient algorithm is established to achieve resilience and attack detection of the remote monitoring system. Finally, it is shown through several case studies on a water supply distribution system that the proposed methods can provide quantitative analysis and evaluation of the potential consequences of various attacks on the remote monitoring system.
       
  • A New Algorithm for Positive Influence Maximization in Signed Networks
    • Abstract: Publication date: Available online 5 November 2019Source: Information SciencesAuthor(s): Weijia Ju, Ling Chen, Bin Li, Wei Liu, Jun Sheng, Yuwei Wang With the rapid development of online social networks, the problem of influence maximization (IM) has attracted much attention from researchers and has been applied in many areas such as marketing and finance. Since positive and negative relations may exist between individuals in social networks, the problem of influence maximization in signed networks has a wide range of applications. This paper presents an efficient algorithm for positive influence maximization in signed networks in the independent cascade model. First, we propose an independent path-based algorithm to compute the activation probabilities between the node pairs. Based on the activation probability, we define a propagation increment function to avoid simulating the influence spreading for selecting candidate seed nodes. We present an algorithm to select the seed nodes to obtain the largest positive influence spreading in the signed network. Empirical results in social networks show that our algorithm can have wider positive influence spreading than other methods.
       
  • ProUM: Projection-Based Utility Mining on Sequence Data
    • Abstract: Publication date: Available online 4 November 2019Source: Information SciencesAuthor(s): Wensheng Gan, Jerry Chun-Wei Lin, Jiexiong Zhang, Han-Chieh Chao, Hamido Fujita, Philip S. Yu Utility is an important concept in Economics. A variety of applications consider utility in real-life situations, which has lead to the emergence of utility-oriented mining (also called utility mining) in the recent decade. Utility mining has attracted a great amount of attention, but most of the existing studies have been developed to deal with itemset-based data. Time-ordered sequence data is more commonly seen in real-world situations, which is different from itemset-based data. Since they are time-consuming and require large amount of memory usage, current utility mining algorithms still have limitations when dealing with sequence data. In addition, the mining efficiency of utility mining on sequence data still needs to be improved, especially for long sequences or when there is a low minimum utility threshold. In this paper, we propose an efficient Projection-based Utility Mining (ProUM) approach to discover high-utility sequential patterns from sequence data. The utility-array structure is designed to store the necessary information of the sequence-order and utility. ProUM can significantly improve the mining efficiency by utilizing the projection technique in generating utility-array, and it effectively reduces the memory consumption. Furthermore, a new upper bound named sequence extension utility is proposed and several pruning strategies are further applied to improve the efficiency of ProUM. By taking utility theory into account, the derived high-utility sequential patterns have more insightful and interesting information than other kinds of patterns. Experimental results showed that the proposed ProUM algorithm significantly outperformed the state-of-the-art algorithms in terms of execution time, memory usage, and scalability.
       
  • Multifactorial Optimization via Explicit Multipopulation Evolutionary
           Framework
    • Abstract: Publication date: Available online 2 November 2019Source: Information SciencesAuthor(s): Genghui Li, Qiuzhen Lin, Weifeng Gao Multifactorial Optimization (MFO) has attracted considerable attention in the community of evolutionary computation, which aims to deal with multiple optimization tasks simultaneously by information transfer. Unfortunately, information transfer may cause both positive and negative effects. To address this issue, this paper exploits an explicit multipopulation evolutionary framework (MPEF) to intelligently take advantage of positive information transfer and effectively reduce negative information transfer. In MPEF, each task possesses an independent population and has its random mating probability for exploiting the information of other tasks. Moreover, the random mating probability of each task is adjusted adaptively. The benefits of using MPEF are twofold. 1) Various well-developed search engines can be easily embedded into MPEF for solving the single task of multifactorial optimization problems. 2) The positive information transfer can be exploited. Meanwhile, negative information transfer can be prevented. A multifactorial evolutionary algorithm (named MFMP) is realized as an instance by embedding a well-designed search engine into MPEF. The experimental results on some MFO benchmark problems demonstrate the advantage of MFMP over some state-of-the-art algorithms. Moreover, MFMP is also successfully employed to solve the spread spectrum radar polyphase code design (SSRPCD) problem.
       
  • Spectral based hypothesis testing for community detection in complex
           networks
    • Abstract: Publication date: Available online 2 November 2019Source: Information SciencesAuthor(s): Zhishan Dong, Shuangshuang Wang, Qun Liu Network analysis is one of the most important branches in modern science, it has brought great advances which help us better understanding complex systems. Recently, detecting community structure within networks has played a more and more critical role in network analysis, due to the fact that it has many crucial applications in a wide range of disciplines, such as sociology, biology, computer science, and other disciplines which can be represented as graphs, hence the problem of detecting communities in networks has attracted a lot of attention from researchers in different areas. However, most of existing algorithms and approaches are built on an assumption that the number of communities in a network is prior known, whereas in many cases, we do not know too much information about this vital quantity. In this work, by fitting networks with stochastic block model, we put forward a novel hypothesis testing framework which can automatically determine the number of communities in various networks. By combining our hypothesis testing method with a motif based clustering approach, we design a recursive bipartitioning algorithm which can fast detect community structure in simulated networks, as well as various real networks.
       
  • Incremental route inference from low-sampling GPS data: an opportunistic
           approach to online map matching
    • Abstract: Publication date: Available online 2 November 2019Source: Information SciencesAuthor(s): Linbo Luo, Xiangting Hou, Wentong Cai, Bin Guo With the surging of smart device sensing and mobile networking, GPS data has been widely available for identifying vehicle position and route on the road map. For many real-time applications, such as traffic sensing and route recommendation, it is critical to immediately infer travelling route with incoming GPS data. In this paper, an opportunistic approach to online map matching is proposed to incrementally infer routes from low-sampling GPS data with low output latency. Unlike the hidden Markov model (HMM)-based approach, which often experiences certain delay between the GPS observation and inference, our algorithm can produce immediate inference when a new GPS point becomes available. Furthermore, a rollback mechanism is provided to correct the already inferred route when some abnormal situations are detected during the opportunistic inference process. We evaluate the proposed algorithm using real dataset of GPS trajectories over 100 cities around the world. Experimental results show that our algorithm is better than, or at least comparable to the state-of-the-art algorithms in terms of inference accuracy. More importantly, our algorithm can yield much shorter output latency and require less execution time, which is critical for many real-time navigation applications and location-based services.
       
  • Distributed Nonlinear Kalman Filter with Communication Protocol
    • Abstract: Publication date: Available online 2 November 2019Source: Information SciencesAuthor(s): Hilton Tnunay, Zhenhong Li, Zhengtao Ding This paper proposes an optimal design of the general distributed nonlinear Kalman-based filtering algorithm to tackle the discrete-time estimation problem with noisy communication networks. The algorithm extends the Kalman filter by enabling it to predict the noisy communication data and fuse it with the received neighboring information to produce a posterior estimate value. In the prediction step, the unscented transformations of the estimate values and covariances originated in the Unscented Kalman Filter (UKF) are exploited. In the update step, a communication protocol is appended to the posterior estimator, which consequently leads to a modified posterior error covariance containing the covariance of the communication term with its communication gain. Both Kalman and communication gains are then optimised to collectively minimise the mean-squared estimation error. Afterwards, stochastic stability analysis is performed to guarantee its exponential boundedness. To exemplify the performance, this algorithm is applied to a group of robots in a sensor network assigned to estimate an unknown information distribution over an area in the optimal coverage control problem. Comparative numerical experiments finally verify the effectiveness of our design.
       
  • Evolutionary many-objective assembly of cloud services via angle and
           adversarial direction driven search
    • Abstract: Publication date: Available online 2 November 2019Source: Information SciencesAuthor(s): Jiajun Zhou, Liang Gao, Xifan Yao, Chunjiang Zhang, Felix T.S. Chan, Yingzi Lin Cloud service composition (CSC) is an effective way to carry out large-scale complicated applications by the ensemble of existing individual services. Each service typically involves several Quality of Service (QoS) criteria contracted for non-functional aspects like time or price, among others, which greatly influence the overall performance of the resulting applications. Service composition approaches have emerged as an important technique in leveraging the quality of composite service efficiently and have attracted significant attention. However, most existing proposals ignore the many-objective nature of CSC and consider up to three objectives, the optimization of diverse QoS aspects of CSC from a many-objective perspective (at least four) still lacks. On another aspect, due to the rapid growth of nondominated solutions in high-dimensional objective spaces, the traditional multi-objective optimization algorithms are usually not capable of handling problems possessing many objectives. To address the above issue, we develop an angle and adversarial direction based optimizer for many-objective CSC scenarios, which evolves a number of subpopulations with adversarial search directions in a parallel paradigm. Additionally, vector angle based selection criterion, which adaptively captures beacon individuals, is utilized to diversify the population. Extensive experiments are carried out on a series of CSC instances utilizing synthetic datasets and the results show that our proposition is competitive and has better versatility compared with the state-of-the-art.
       
  • Sparse unmixing of hyperspectral data with bandwise model
    • Abstract: Publication date: Available online 1 November 2019Source: Information SciencesAuthor(s): Chang Li, Yu Liu, Juan Cheng, Rencheng Song, Jiayi Ma, Chenhong Sui, Xun Chen Sparse unmixing has long been a hot research topic in the area of hyperspectral image (HSI) analysis. Most of the traditional sparse unmixing methods usually assume to only take the Gaussian noise into consideration. However, there are also other types of noise in real HSI, i.e., impulse noise, stripes, dead lines and so on. In addition, the intensity of Gaussian noise is usually different for each band of real HSI. To this end, we propose a novel sparse unmixing method with the bandwise model (SUBM) to address the above mentioned problems simultaneously. Besides, the alternative direction method of multipliers (ADMM) is adopted for solving the proposed SUBM. Moreover, we conduct extensive experiments on synthetic and real datasets to demonstrate effectiveness of the proposed sparse unmixing method under the bandwise model.
       
  • Design of an Interval Type-2 Fuzzy Model with Justifiable Uncertainty
    • Abstract: Publication date: Available online 1 November 2019Source: Information SciencesAuthor(s): Juan E. Moreno, Mauricio A. Sanchez, Olivia Mendoza, Antonio Rodríguez-Díaz, Oscar Castillo, Patricia Melin, Juan R. Castro Throughout previous design proposals of Interval Type-2 Fuzzy Logic Systems most of the research work concentrates on optimal design to best fit data behavior and rarely focus on the inner model essence of Type-2 Fuzzy Systems, which is uncertainty. In this way, failing to focus on this key aspect, which is how much uncertainty exists within the model to better represent the data. In this paper a design methodology for a Mamdani based Interval Type-2 Fuzzy Logic System (MAM-IT2FLS) with Center-Of-Sets defuzzification is presented, using descriptive statistics and granular computing theory to better define the limits of uncertainty within the Interval Type-2 Membership Functions (IT2MF) as extracted from available data. This allows us to justify the uncertainty within the entire Type-2 Fuzzy Logic model, as well as to create the fuzzy model using FCM grouping and to compute IT2MF parameters from MAM-IT2FLS rules using simple steps. This is unlike hybrid learning models with Back-Propagation that adjust IT2MF parameters with gradient based numeric optimization algorithms which are time efficient but unstable for convergence, and evolutionary computation with robust convergence and slow learning time. Experimentation is carried out with six regression benchmark datasets, measuring RMSE and R2 in order to evaluate the performance of the proposed methodology whilst maintaining justifiable uncertainty in its model.
       
  • A Spatiotemporal Attention Mechanism-based Model for Multi-step Citywide
           Passenger Demand Prediction
    • Abstract: Publication date: Available online 1 November 2019Source: Information SciencesAuthor(s): Yirong Zhou, Jun Li, Hao Chen, Ye Wu, Jiangjiang Wu, Luo Chen In taxi dispatch systems, predicting citywide passenger pickup/dropoff demand is indispensable for developing effective taxi distribution and scheduling strategies to resolve the demand-service mismatch. Compared with predicting next-step only, predicting multiple steps is preferable since it can provide a long term view, thus preventing short-sighted strategies. However, multi-step citywide passenger demand prediction (MsCPDP) is challenging due to the complicated spatiotemporal correlations in the distribution of passenger demand and the lack of ground truth from pre-steps for the prediction of subsequent steps. In this paper, a deep-learning-based prediction model with spatiotemporal attention mechanism is proposed for MsCPDP. The model, called ST-Attn, follows the general encoder-decoder framework for modelling sequential data but adopts a multiple-output strategy without recurrent neural network units. The spatiotemporal attention mechanism learns to determine the focus on those parts of the city at certain periods that are more relevant to the passenger demand in the predicted region and time period. In addition, a pre-predicted result calculated by spatiotemporal kernel density estimation is fed to ST-Attn, which provides a reference for further accurate prediction. Experiments on three real-world datasets are carried out to verify ST-Attn’s performance, and the results show that ST-Attn outperforms the baselines in terms of MsCPDP.
       
  • An Uncertainty-Aware Computational Trust Model Considering the
           Co-existence of Trust and Distrust in Social Networks
    • Abstract: Publication date: Available online 1 November 2019Source: Information SciencesAuthor(s): Nastaran Hakimi Aghdam, Mehrdad Ashtiani, Mohammad Abdollahi Azgomi There are few trust models capable of incorporating the co-existence of trust and distrust as distinct concepts. In this regard, most of the existing trust models implicitly use distrust parameters to refine and calculate trust values. However, recent studies have indicated that trust and distrust are two distinct but co-existing concepts. In other words, although trust and distrust are constructed based on different characteristics, they can be used together in the decision making and recommendation processes. In this paper, we present a trust-distrust model for social networks considering subjective and objective characteristics of trust and distrust simultaneously. Competence, honesty, satisfaction, similarity, motivation, availability, tendency to be trusted, the existence of long-term connection/friendship, and centrality are the trustworthiness characteristics covered by the model. Also, surprisal, dishonesty, dissatisfaction, conflict degree, account lifetime, and sudden changes in the number of friends, likes, and comments are the distrust characteristics considered by the model. The proposed model takes into account the uncertainty, sharpness, and vagueness of the beliefs by using subjective logic. The results of the conducted evaluations demonstrate that the proposed model is highly accurate in the decision-making process and has a 90% accuracy in calculating the trust and distrust. We have also compared the results with other similar approaches, by which the proposed model showed a 34% improvement.
       
  • Block Change Learning for Knowledge Distillation
    • Abstract: Publication date: Available online 1 November 2019Source: Information SciencesAuthor(s): Hyunguk Choi, Younkwan Lee, Kin Choong Yow, Moongu Jeon Deep neural networks perform well but require high-performance hardware for their use in real-world environments. Knowledge distillation is a simple method for improving the performance of a small network by using the knowledge of a large complex network. Small and large networks are referred to as student and teacher models, respectively. Previous knowledge distillation approaches perform well in a relatively small teacher network (20 ∼ 30 layers) but poorly in large teacher networks (50 layers). Here, we propose an approach called block change learning that performs local and global knowledge distillation by changing blocks comprised of layers. The method focuses on the knowledge transfer without losing information in a large teacher model, as the approach considers intra-relationships between layers using local knowledge distillation and inter-relationships between corresponding blocks. The results are demonstrated this approach as superior to state-of-the-art methods using feature extraction datasets (Market1501 and DukeMTMC-relD) and object classification datasets (CIFAR-100 and Caltech256). Furthermore, we showed that the performance of the proposed approach was superior to that of a fine-tuning approach using pretrained models.
       
  • Minimizing Rumor Influence in Multiplex Online Social Networks Based on
           Human Individual and Social Behaviors
    • Abstract: Publication date: Available online 1 November 2019Source: Information SciencesAuthor(s): Hosni Adil Imad Eddine, Kan Li, Sadique Ahmad With the growing popularity of online social networks, an environment has been set up that can spread rumors in a faster and wider manner than ever before, which can have widespread repercussions on society. Nowadays, individuals are joining multiple online social networks and rumors simultaneously propagating amongst them, thereby creating a new dimension to the problem of rumor propagation. Motivated by these facts, this paper attempts to address the rumor influence minimization in multiplex online social networks. In this work, we consider modeling the propagation process of such fictitious information as a significant step toward minimizing its influence. Thus, we analyze the individual and social behaviors in social networks; subsequently, we propose a novel rumor diffusion model, named the HISBmodel. In this model, we propose a formulation of an individual behavior towards a rumor analog to damped harmonic motion. Following this, the opinions of individuals in the propagation process are incorporated. Furthermore, the rules of rumor transmission between individuals in multiplex networks are incorporated by considering individual and social behaviors. Further, we present the HISBmodel propagation process that describes the spread of rumors in multiplex online social networks. Based on this model, we propose a truth campaign strategy in minimizing the influence of rumors in multiplex online social networks from the perspective of network inference and by exploiting the survival theory. This strategy selects the most influential nodes as soon as the rumor is detected and launches a truth campaign to raise awareness against it, so as to prevent the influence of rumors. Accordingly, we propose a greedy algorithm based on the likelihood principle, which guarantees an approximation within 63% of the optimal solution. Systematically, experiments have been conducted on real single networks crawled from Twitter, Facebook, and Slashdot as well as on multiplex networks of real online social networks (Facebook, Twitter, and YouTube). First, the results indicate the HISBmodel can reproduce all the trends of real-world rumor propagation more realistically than the models presented in the literature. Moreover, the simulations illustrate that the proposed model highlights the impact of human factors accurately in accordance with the literature. Second, compared to the methods in the literature, the experiments prove the efficiency of our strategy in minimizing the influence of rumors in the cases of single network and multiplex social network propagation. The results prove that the proposed method can capture the dynamic propagation process of the rumor and select the target nodes more accurately in order to minimize the influence of rumors.
       
  • Erratum to “An Efficient Linear Programming Based Method for the
           Influence Maximization Problem in Social Networks” [Information Sciences
           503C (201911) 589–605]
    • Abstract: Publication date: February 2020Source: Information Sciences, Volume 511Author(s): Evren Guney
       
  • Inference attacks on genomic privacy with an improved HMM and an RCNN
           model for unrelated individuals
    • Abstract: Publication date: Available online 1 October 2019Source: Information SciencesAuthor(s): Hongfa Ding, Youliang Tian, Changgen Peng, Youshan Zhang, Shuwen Xiang In recent years, the collection of large-scale genomic data for individuals has become feasible and affordable. Concurrently, several practical attacks targeting genome re-identification and genotype inference have emerged to threaten the confidentiality of genomic data sharing, leading to security and privacy concerns regarding genomic data. The authors have shown that this problem can be even worse in this paper. Specifically, two possible large-scale genotype inference attack stretegies for nonrelatives have exposed. One is based on an improved hidden Markov model (iHMM), and the other is based on a regressive convolutional neural network (RCNN). By using a genomic privacy metric combining the attacker’s incorrectness, the attacker’s uncertainty, and the genomic privacy loss of the victims, it is shown that with these atrategies, the attack can be significantly more severe than those reported previously. It is also shown that machine learning can be applied to empower large-scale inference attacks against genomic privacy.
       
  • A Tree-based Neural Network Model for Biomedical Event Trigger Detection
    • Abstract: Publication date: Available online 30 September 2019Source: Information SciencesAuthor(s): Hao Fei, Yafeng Ren, Donghong Ji Biomedical event trigger detection is a heated research topic since its important role in biomedical event extraction. Previous studies show that syntactic features are very crucial for the task. However, existing methods largely focus on traditional statistical models, and usually capture syntactic features by extracting a set of manually-crafted features based on dependency tree. This limits the performance of the task. In this paper, we propose a tree-based neural network model, which can automatically learn syntactic features from dependency tree for trigger detection. Specifically, we use a recursive neural network to represent whole dependency tree globally, to better incorporate dependency-based syntax information. Results on MLEE and BioNLP-09 datasets show that the proposed model achieves 80.28% and 73.24% F1 score, respectively, outperforming traditional statistical models and neural baseline systems.
       
  • Auto-weighted Multi-view Co-clustering with Bipartite Graphs
    • Abstract: Publication date: Available online 30 September 2019Source: Information SciencesAuthor(s): Shudong Huang, Zenglin Xu, Ivor W. Tsang, Zhao Kang Co-clustering aims to explore coherent patterns by simultaneously clustering samples and features of data. Several co-clustering methods have been proposed in the past decades. However, in real-world applications, datasets are often with multiple modalities or composed of multiple representations (i.e., views), which provide different yet complementary information. Hence, it is essential to develop multi-view co-clustering models to solve the multi-view application problems. In this paper, a novel multi-view co-clustering method based on bipartite graphs is proposed. To make use of the duality between samples and features of multi-view data, a bipartite graph for each view is constructed such that the co-occurring structure of data can be extracted. The key point of utilizing the bipartite graphs to deal with the multi-view co-clustering task is to reasonably integrate these bipartite graphs and obtain an optimal consensus one. As for this point, the proposed method can learn an optimal weight for each bipartite graph automatically without introducing an additive parameter as previous methods do. Furthermore, an efficient algorithm is proposed to optimize this model with theoretically guaranteed convergence. Extensive experimental results on both toy data and several benchmark datasets have demonstrated the effectiveness of the proposed model.
       
  • Calculus for linearly correlated fuzzy function using Fréchet
           Derivative and Riemann Integral
    • Abstract: Publication date: Available online 30 September 2019Source: Information SciencesAuthor(s): Francielle Santo Pedro, Estevão Esmi, Laécio Carvalho de Barros In this manuscript we study integration and derivative theories for interactive fuzzy processes. These theories are based on the Fréchet derivative and the Riemann integral. In addition, we present a connection between these two theories, i.e., some problems may be formulated in both ways. We establish the fundamental theorem of calculus, theorem of existence and the local uniqueness of the solution of fuzzy differential equations and some techniques to solve fuzzy initial value problems. To illustrate the usefulness of the developed theory, we investigate the radioactive decay model.
       
  • Vector Partitioning Quantization Utilizing K-means Clustering for Physical
           Layer Secret Key Generation
    • Abstract: Publication date: Available online 30 September 2019Source: Information SciencesAuthor(s): Qingqing Han, Jingmei Liu, Zhiwei Shen, Jingwei Liu, Fengkui Gong Most existing key generation schemes cannot achieve the optimal performance because the quantization algorithm cannot make full use of channel information, some schemes use amplitude quantization and others use phase quantization. This paper proposes the concept of vector partitioning quantization (VPQ), and several new quantization algorithms are obtained further, which can utilize amplitude and phase at the same time. First, the traditional amplitude quantization (TAQ) and the traditional phase quantization (TPQ) are reviewed in detail. Second, the regular vector partitioning quantization (RVPQ) is proposed based on TAQ and TPQ, and the error probabilities of RVPQ is deduced, moreover, the concept of optimal RVPQ is presented. Then, three irregular vector partitioning quantization (IVPQ) algorithms are proposed which utilizes K-means clustering, including basic K-means quantization (BKQ), lossy K-means quantization (LKQ) and compensation K-means quantization (CKQ). The three K-means quantization algorithms can not only reduce the quantization error rate greatly, but also ensure the confidentiality of key generation. To overcome the problem of weak uniformity caused by K-means, we put forward the balance mechanism (BM) and apply it to three K-means quantization algorithms. Finally, the simulation results show the VPQ algorithms proposed are superior to previous methods.
       
  • Designing Privacy-aware Internet of Things Applications
    • Abstract: Publication date: Available online 28 September 2019Source: Information SciencesAuthor(s): Charith Perera, Mahmoud Barhamgi, Arosha K. Bandara, Muhammad Ajmal, Blaine Price, Bashar Nuseibeh Internet of Things (IoT) applications typically collect and analyse personal data that can be used to derive sensitive information about individuals. However, thus far, privacy concerns have not been explicitly considered in software engineering processes when designing IoT applications. The advent of behaviour driven security mechanisms, failing to address privacy concerns in the design of IoT applications can have security implications. In this paper, we explore how a Privacy-by-Design (PbD) framework, formulated as a set of guidelines, can help software engineers integrate data privacy considerations into the design of IoT applications. We studied the utility of this PbD framework by studying how software engineers use it to design IoT applications. We also explore the challenges in using the set of guidelines to influence the IoT applications design process. In addition to highlighting the benefits of having a PbD framework to make privacy features explicit during the design of IoT applications, our studies also surfaced a number of challenges associated with the approach. A key finding of our research is that the PbD framework significantly increases both novice and expert software engineers’ ability to design privacy into IoT applications.
       
  • Robust Event-Triggered Reliable Control for T-S Fuzzy Uncertain Systems
           via Weighted Based Inequality
    • Abstract: Publication date: Available online 27 September 2019Source: Information SciencesAuthor(s): G. Nagamani, Young Hoon Joo, G. Soundararajan, Reza Mohajerpoor This paper is concerned with the design of event-triggered reliable control scheme for a class of uncertain Takagi-Sugeno (T-S) fuzzy nonlinear systems involving network transmission delays. Moreover, the designed fuzzy controller is reliable in the sense that the stability and the satisfactory performance of the closed-loop system are achieved not only under normal operation but in the presence of some actuator faults as well. For this notion, a reliable event-trigger scheme is designed for the nonlinear system in transmitting the sampled state information to the controller. By employing Lyapunov-Kraskovskii functional (LKF) and by using the weighted integral inequality, the exponential stability conditions are derived in the form of linear matrix inequalities (LMIs). By solving the proposed LMI conditions the triggering and control gain matrices are obtained via the computational toolbox. Finally, the effectiveness and the less conservativeness of the proposed robust event-triggered controller have been illustrated via some real control systems such as mass-spring-damper physical system, flexible-joint robot arm and variable speed wind turbine system.
       
  • A probabilistic view on semilinear copulas
    • Abstract: Publication date: Available online 27 September 2019Source: Information SciencesAuthor(s): Henrik Sloot, Matthias Scherer This article advances the theory on multivariate upper semilinear copulas. Probabilistic features of this class are discussed and three subclasses are investigated in detail. The first subclass consists of upper semilinear copulas whose survival copulas are generalised Marshall–Olkin copulas. The second subclass is defined in that they possess identical multivariate diagonals. The third subclass is a family of extendible upper semilinear copulas. Stochastic models and analytical characterisation theorems are derived for each of these subclasses.
       
  • Adaptive decentralized output feedback PI tracking control design for
           uncertain interconnected nonlinear systems with input quantization
    • Abstract: Publication date: Available online 27 September 2019Source: Information SciencesAuthor(s): Haibin Sun, Guangdeng Zong, C.L. Philip Chen In this paper, the problem of adaptive decentralized proportional-integral (PI) tracking control is investigated for a class of interconnected nonlinear systems with input quantization and unknown functions, where the interconnection terms are bounded by completely unknown functions. By designing an input-driven filter, the unknown states are estimated and then an adaptive decentralized output feedback PI tracking controller is constructed via the backstepping method and neural network technique. The stability of the closed-loop system is addressed based on the Lyapunov function technique plus graph theory, and all the signals in the closed-loop system are uniformly ultimately bounded. Finally, simulation results are utilized to demonstrate the effectiveness of the proposed method.
       
  • Bipartite Consensus for Networked Robotic Systems with Quantized-data
           Interactions
    • Abstract: Publication date: Available online 26 September 2019Source: Information SciencesAuthor(s): Teng-Fei Ding, Ming-Feng Ge, Cai-Hua Xiong, Ju H. Park The bipartite consensus problem of networked robotic systems (NRSs) with parametric uncertainties, input disturbances, and quantized-data interactions is addressed in this paper. Some novel distributed estimator-based control algorithms are designed to guarantee that all controlled robots can eventually reach bipartite consensus or their states asymptotically converge to the origin. By employing the Lyapunov argument and nonsmooth analysis theory, several sufficient criteria on control parameters for stabilizing the closed-loop systems and solving the aforementioned problems are provided. Finally, simulation examples are presented to illustrate the proposed algorithms.
       
  • CCFR2: A More Efficient Cooperative Co-Evolutionary Framework for
           Large-Scale Global Optimization
    • Abstract: Publication date: Available online 26 September 2019Source: Information SciencesAuthor(s): Ming Yang, Aimin Zhou, Changhe Li, Jing Guan, Xuesong Yan Cooperative co-evolution (CC) is an explicit means of divide-and-conquer strategy in evolutionary algorithms for solving large-scale optimization problems. The subcomponents generated by CC may have different characteristics. When optimizing the subcomponents, the settings of the subpopulations should take the characteristics into account. CCFR is a previously published CC framework which allocates computational resources among subpopulations according to the contributions of subpopulations to the improvement of the best overall objective value. In this paper, we propose an improved version of CCFR named CCFR2, which can specify unequal-sized subpopulations for optimizing different subcomponents of variables. CCFR2 computes the average improvement of the best overall objective value per fitness evaluation as the contribution of a subpopulation, which considers the subpopulation size in the contribution computation. A control parameter is adopted by CCFR2 to balance the effects of the historical and real-time improvements of the best overall objective value on the contribution computation. Compared with CCFR, CCFR2 is able to save computational resources from obtaining the best overall solution before the evolution starts and evaluating individuals in co-evolutionary cycles. Our experimental results and analysis suggest that CCFR2 improves the performance of CCFR and is a more efficient CC framework for solving large-scale optimization problems.
       
  • Toward Optimal Participant Decisions with Voting-based Incentive Model for
           Crowd Sensing
    • Abstract: Publication date: Available online 26 September 2019Source: Information SciencesAuthor(s): Nan Jiang, Dong Xu, Jie Zhou, Hongyang Yan, Tao Wan, Jiaqi Zheng With the rapid development of crowd sensing in sensing applications, excellent incentive mechanisms are playing an increasingly important role. However, most existing solutions do not fully consider the ability of participants to perform tasks, the degree to which they complete tasks, or the credibility of the task sensing results. In this paper, we aim to develop an incentive model based on voting mechanism for crowd sensing(abbreviated as CIBV), which includes three algorithms. The first is a participant decision algorithm (PDA) that adopts a reverse auction model and comprehensively considers candidate execution capability; the second is the budget balance and extra reward algorithm (BBER); the third is the evaluate algorithm (EA) to be applied at the end of sensing tasks. Compared with previous work, the experimental results show that in our proposed CIBV model, each task is performed by multiple participants, and each participant can perform multiple tasks, our model can greatly improve the participants’ execution ability value and provide the platform with the ability to control the process of selecting participants.
       
  • Incentive Mechanism for the Listing Item Task in Crowdsourcing
    • Abstract: Publication date: Available online 26 September 2019Source: Information SciencesAuthor(s): Shaofei Wang, Depeng Dang Crowdsourcing is a new strategy of leveraging intelligence from a large number of workers to complete tasks. An incentive mechanism is an effective way for improving the quality of answers in crowdsourcing. However, a special but common type of crowdsourcing task, called listing item task, has not been fully investigated. In this paper, we focus on the incentive mechanism for this listing item task. In particular, we first provide a formal definition of this task. Then, we propose an effective incentive mechanism considering both the precision and recall of the answers. Next, we prove that the proposed mechanism is incentive-compatible and satisfies no free lunch criterion. Finally, we conduct a series of experiments on our crowdsourcing platform CrowdKnow and a public platform ZhiDao. The experimental results demonstrate that our incentive mechanism achieves a remarkable improvement for listing item tasks compared with other related mechanisms.
       
  • A projection-based regret theory method for multi-attribute decision
           making under interval type-2 fuzzy sets environment
    • Abstract: Publication date: Available online 26 September 2019Source: Information SciencesAuthor(s): Huidong Wang, Xiaohong Pan, Jun Yan, Jinli Yao, Shifan He Interval type-2 fuzzy sets (IT2FSs) can provide more flexibility than type-1 fuzzy sets (T1FSs) for depicting uncertain information, and multi-attribute decision making (MADM) problems with interval type-2 fuzzy information have received increasing attention. A new projection-based regret theory method is proposed to solve MADM problems under IT2FSs environments. First, a projection model of IT2FSs is defined that takes both the distance and angle information into consideration. Second, integrating the proposed projection model with regret theory, new utility and regret-rejoice functions are developed, respectively. Finally, a case study is provided to demonstrate the effectiveness of the proposed method. Sensitivity analysis shows the stability of the proposed method, and the ranking order does not change with different parameters. Comparisons are made with existing approaches to illustrate the advantage of the proposed method in reflecting decision makers’ psychological factors.
       
  • Parallel Hierarchical Architecturesfor Efficient Consensus Clusteringon
           Big Multimedia Cluster Ensembles
    • Abstract: Publication date: Available online 26 September 2019Source: Information SciencesAuthor(s): Xavier Sevillano, Joan Claudi Socoró, Francesc Alías Consensus clustering is a useful tool for robust or distributed clustering applications. However, given the fact that time complexities of the consensus functions scale linearly or quadratically with the number of combined clusterings, execution can be slow or even impossible when operating on big cluster ensembles, a situation encountered when we pursue robust multimedia data clustering. This work introduces hierarchical consensus architectures, an inherently parallel approach based on the divide-and-conquer strategy for computationally efficient consensus clustering, in a bid to make faster, more effective consensus clustering possible in big multimedia cluster ensemble scenarios. Moreover, we define a specific implementation of hierarchical architectures, including a theoretical analysis of its fully parallel implementation computational complexity. In experiments conducted on unimodal and multimedia data sets involving small and big cluster ensembles, we find parallel hierarchical consensus architectures variants perform faster than traditional flat consensus in 75% of the experiments on small cluster ensembles, a percentage that rises to 100% on unimodal and multimedia big cluster ensembles, achieving an average speedup ratio of 30.5. Moreover, depending on the consensus function employed, the quality of the obtained consensus partitions ensures robust clustering results.
       
  • Robust reversible data hiding scheme based on two-layer embedding strategy
    • Abstract: Publication date: Available online 26 September 2019Source: Information SciencesAuthor(s): Rajeev Kumar, Ki-Hyun Jung Robust reversible data hiding (RRDH) prevents the hidden secret information from unintentional modifications. This paper presents a novel RRDH scheme based on two-layer embedding with reduced capacity-distortion trade-off. The proposed scheme first decomposes the image into two planes namely higher significant bit (HSB) and least significant bit (LSB) planes and then employs prediction error expansion (PEE) to embed the secret data into the HSB plane. The high correlation of HSB plane helps in achieving high embedding capacity. Further, non-malicious attacks such as Joint Photographic Experts Group (JPEG) compression which usually changes the LSBs, will not cause any disturbance to the main contents of original image and the hidden secret data. The experimental results show that the proposed scheme has superior performance than the previous works.
       
  • On state estimation for nonlinear dynamical networks with random sensor
           delays and coupling strength under event-based communication mechanism
    • Abstract: Publication date: Available online 25 September 2019Source: Information SciencesAuthor(s): Jun Hu, Guo-Ping Liu, Hongxu Zhang, Hongjian Liu This paper is concerned with the optimized state estimation problem for nonlinear dynamical networks subject to random coupling strength (RCS) and random sensor delays (RSDs) under the event-triggered communication criterion. Firstly, a set of random variables obeying uniform distribution over certain interval is introduced to reflect the stochasticity of the elements of coupling configuration. Furthermore, a series of Bernoulli distributed random variables is introduced to characterize the phenomenon of RSDs, where the inaccuracy of the occurrence probability is depicted. Besides, as a special communication protocol, the event-triggered communication scheme is employed to avoid the potential data collision. Subsequently, a new robust state estimation algorithm is provided for addressed nonlinear dynamical networks by fully taking the available information of RCS and RSDs into account. Moreover, a sufficient condition ensuring the boundedness of state estimation error is presented. Finally, some simulations are utilized to demonstrate the usefulness of the optimized estimation technique proposed in this paper.
       
  • Memory based self-adaptive sampling for noisy multi-objective optimization
    • Abstract: Publication date: Available online 25 September 2019Source: Information SciencesAuthor(s): Pratyusha Rakshit The paper proposes a novel strategy to adapt sample size of population members of a multi-objective optimization (MOO) problem, where the objective surface is contaminated with noise. The sample size, used for periodic fitness evaluation of a solution, is adapted based on the fitness variance of a sub-population in its neighborhood and is referred to as local neighborhood fitness variance (LNFV). The constraint of selecting accurate functional relationship between sample size and LNFV is surmounted here by employing a novel memory-based sample size adaptation policy. In the early exploration phase of a MOO, the policy memorizes the success or failure of sample sizes assigned to solutions with specific LNFVs. These success and failure history are later utilized to guide solutions of future generations to carefully select sample sizes based on their individual LNFVs. Experiments undertaken disclose the superiority of the proposed realization to the existing counterparts and the state-of-the-art techniques. The proposed algorithms have also been applied on a multi-robot box-pushing problem where the sensory data of twin robots are contaminated with noise. Experimental results here too reveal the efficiency of the proposed realizations in terms of minimization of execution time and energy consumed by the twin robots.
       
  • Flexible Attribute-based Proxy Re-encryption for Efficient Data Sharing
    • Abstract: Publication date: Available online 24 September 2019Source: Information SciencesAuthor(s): Hua Deng, Zheng Qin, Qianhong Wu, Zhenyu Guan, Yunya Zhou An increasing number of people are sharing their data through third-party platforms. Attribute-based encryption (ABE) is a promising primitive that allows enforcing fine-grained access control on the data to be shared. An issue in ABE is that a priori access policies should be determined during the system setup or encryption phase, but these policies will become obsolete over time. Another issue is that the decryption of ABE generally requires complicated and expensive computations, which may be unaffordable for resource-limited users (e.g., mobile-device users). To address these issues, we propose a new paradigm called hybrid attribute-based proxy re-encryption (HAPRE). In HAPRE, a semitrusted proxy can be authorized to convert ciphertexts of an ABE scheme into ciphertexts of an identity-based encryption (IBE) scheme without letting the proxy know the underlying messages. With these features, HAPRE enables resource-limited users to efficiently access the data previously encrypted by ABE. We construct two HAPRE schemes by utilizing a compact IBE scheme and a key rerandomization technique, and then we prove that the schemes are semantically secure and collusion resistant. Theoretical and experimental analyses demonstrate the efficiency of the HAPRE schemes.
       
  • Shape-Optimizing Mesh Warping Method for Stereoscopic Panorama Stitching
    • Abstract: Publication date: Available online 24 September 2019Source: Information SciencesAuthor(s): Weiqing Yan, Guanghui Yue, Jindong Xu, Yanwei Yu, Kai Wang, Chang Tang, Xiangrong Tong In this paper, we propose a novel shape-optimizing mesh warping method for stereoscopic panorama stitching, which aims to resolve shape distortion and unnatural rotation of traditional stitching methods, simultaneously coping with the challenges, misalignment, and stereoscopic inconsistency. Specifically, based on the grid mesh analysis of projective warping, we propose a differential warping method by gradually changing the inclination angle of each mesh line in non-overlapping regions of the image to reduce shape distortion and unnatural rotation. Furthermore, an extended moving direct linear transformation method is proposed to effectively and robustly improve alignment accuracy and maintain stereoscopic consistency in multiple stereoscopic images. Finally, a consistent seam based on the matched feature points in the left- and right- view images of a stereoscopic image is designed to blend images and generate a stereoscopic panorama image. Experiments demonstrate that the proposed method has a superior performance compared to previous methods.
       
  • A Multiobjective Multifactorial Optimization Algorithm based on
           Decomposition and Dynamic Resource Allocation Strategy
    • Abstract: Publication date: Available online 24 September 2019Source: Information SciencesAuthor(s): Shuangshuang Yao, Zhiming Dong, Xianpeng Wang, Lei Ren Multiobjective multifactorial optimization (MO-MFO), i.e., multiple multiobjective tasks are simultaneously optimized by a single population, has received considerable attention in recent years. Traditional algorithms for the MO-MFO usually allocate equal computing resources to each task, however, this may not be reasonable due to the fact that different tasks usually have different degrees of difficulty. Motivated by the idea that the limited computing resources should be adaptively allocated to different tasks according to their difficulties, this paper proposes an algorithm for the MO-MFO based on decomposition and dynamic resource allocation strategy (denoted as MFEA/D-DRA). In the MFEA/D-DRA, each multiobjective optimization task is firstly decomposed into a series of single-objective subproblems. Thereafter, a single population is used to evolve all the single-objective subproblems. In the process of evolution, subproblems with fast evolution rate will have the opportunity to get more rewards, i.e., computing resources. The evolution rate is measured by a utility function and updated periodically. Moreover, different multiobjective optimization tasks can communicate with each other according to a random mating probability. Finally, a set of evenly distributed approximate Pareto optimal solutions is obtained for each multiobjective optimization task. The statistical analysis of experimental results illustrates the superiority of the proposed MFEA/D-DRA algorithm on a variety of benchmark MO-MFO problems.
       
  • Clicking position and user posting behavior in online review systems: A
           data-driven agent-based modeling approach
    • Abstract: Publication date: Available online 24 September 2019Source: Information SciencesAuthor(s): Guoyin Jiang, Xiaodong Feng, Wenping Liu, Xingjun Liu In online review systems, a participant's level of knowledge impacts his/her posting behaviors, and an increase in knowledge occurs when the participant reads the reviews posted on the systems. To capture the collective dynamics of posting reviews, we used real-world big data collected over 153 months to drive an agent-based model for replicating the operation process of online review systems. The model explains the effects of clicking position (e.g., on a review webpage's serial list) and the number of items per webpage on posting contributions. Reading reviews from the last webpage only, or from the first webpage and last webpage simultaneously, can promote a greater review volume than reading reviews in other positions. This illustrates that representing primacy (first items) and recency (recent items) within one page simultaneously, or displaying recent items in reverse chronological order, are relatively better strategies for the webpage display of online reviews. The number of items plays a nonlinear moderating role in bridging the clicking position and posting behavior, and we determine the optimal number of items. To effectively establish strategies for webpage design in online review systems, business managers must switch from reliance on experience to reliance on an agent-based model as a decision support system for the formalized webpage design of online review systems.
       
  • DeePr-ESN: A Deep Projection-Encoding Echo-State Network
    • Abstract: Publication date: Available online 24 September 2019Source: Information SciencesAuthor(s): Qianli Ma, Lifeng Shen, Garrison W. Cottrell As a recurrent neural network that requires no training, the reservoir computing (RC) model has attracted widespread attention in the last decade, especially in the context of time series prediction. However, most time series have a multiscale structure, which a single-hidden-layer RC model may have difficulty capturing. In this paper, we propose a novel multiple projection-encoding hierarchical reservoir computing framework called Deep Projection-encoding Echo State Network (DeePr-ESN). The most distinctive feature of our model is its ability to learn multiscale dynamics through stacked ESNs, connected via subspace projections. Specifically, when an input time series is projected into the high-dimensional echo-state space of a reservoir, a subsequent encoding layer (e.g., an autoencoder or PCA) projects the echo-state representations into a lower-dimensional feature space. These representations are the principal components of the echo-state representations, which removes the high frequency components of the representations. These can then be processed by another ESN through random connections. By using projection layers and encoding layers alternately, our DeePr-ESN can provide much more robust generalization performance than previous methods, and also fully takes advantage of the temporal kernel property of ESNs to encode the multiscale dynamics of time series. In our experiments, the DeePr-ESNs outperform both standard ESNs and existing hierarchical reservoir computing models on some artificial and real-world time series prediction tasks.
       
  • Collaborative Linear Manifold Learning for Link Prediction in
           Heterogeneous Networks
    • Abstract: Publication date: Available online 24 September 2019Source: Information SciencesAuthor(s): JiaHui Liu, Xu Jin, YuXiang Hong, Fan Liu, QiXiang Chen, YaLou Huang, MingMing Liu, MaoQiang Xie, FengChi Sun Link prediction in heterogeneous networks aims at predicting missing interactions between pairs of nodes with the help of the topology of the target network and interconnected auxiliary networks. It has attracted considerable attentions from both computer science and bioinformatics communities in the recent years. In this paper, we introduce a novel Collaborative Linear Manifold Learning (CLML) algorithm. It can optimize the consistency of nodes similarities by collaboratively using the manifolds embedded between the target network and the auxiliary network. The experiments on four benchmark datasets have demonstrated the outstanding advantages of CLML, not only in the high prediction performance compared to baseline methods, but also in the capability to predict the unknown interactions in the target networks accurately and effectively.
       
  • A Convolutional Neural-based Learning Classifier System for Detecting
           Database Intrusion via Insider Attack
    • Abstract: Publication date: Available online 24 September 2019Source: Information SciencesAuthor(s): Seok-Jun Bu, Sung-Bae Cho Role-based access control (RBAC) in databases provides a valuable level of abstraction to promote security administration at the business enterprise level. With the capacity for adaptation and learning, machine learning algorithms are suitable for modeling normal data access patterns based on large amounts of data and presenting robust statistical models that are not sensitive to user changes. We propose a convolutional neural-based learning classifier system (CN-LCS) that models the role of queries by combining conventional learning classifier system (LCS) with convolutional neural network (CNN) for a database intrusion detection system based on the RBAC mechanism. The combination of modified Pittsburgh-style LCSs for the optimization of feature selection rules and one-dimensional CNNs for modeling and classification in place of traditional rule generation outperforms other machine learning classifiers on a synthetic query dataset. In order to quantitatively compare the inclusion of rule generation and modeling processes in the CN-LCS, we have conducted 10-fold cross-validation tests and analysis through a paired sampled t-test.
       
  • An Evolutionary Approach for Efficient Prototyping of Large Time Series
           Datasets
    • Abstract: Publication date: Available online 23 September 2019Source: Information SciencesAuthor(s): Pablo Leon-Alcaide, Luis Rodriguez-Benitez, Ester Castillo-Herrera, Juan Moreno-Garcia, Luis Jimenez-Linares We here describe an algorithm based on an evolutionary strategy to find the prototype series of a set of time series, and we use Dynamic Time Warping (DTW) as a distance measure between series, and do not restrict the search space to the series in the set. The problem of calculating the centroid of a set of time series can be addressed as a minimization problem, using genetic algorithms. Our proposal may be considered among the set of non-classical approaches to genetic algorithms, where an individual gene is a candidate time series for being the centroid or representative of the whole set of series. The representation and operators of genetic algorithms are redesigned, in order to generate efficient summaries, the fitness function of each candidate series to be a prototype is approximated, comparing them only with a subset of randomly selected time series from the original dataset. Three areas are looked at in order to assess the goodness of our proposal: the performance of the prototype generated in terms of a fitness function, the consistency of the prototype generation for use in classical grouping algorithms, and its use in classification algorithms based on the nearest prototypes.
       
  • Image Splicing Forgery Detection Combining Coarse to Refined Convolutional
           Neural Network and Adaptive Clustering
    • Abstract: Publication date: Available online 23 September 2019Source: Information SciencesAuthor(s): Bin Xiao, Yang Wei, Xiuli Bi, Weisheng Li, Jianfeng Ma This paper proposes a splicing forgery detection method with two parts: a coarse-to-refined convolutional neural network (C2RNet) and diluted adaptive clustering. The proposed C2RNet cascades a coarse convolutional neural network (C-CNN) and a refined CNN (R-CNN) and extracts the differences in the image properties between un-tampered and tampered regions from image patches with different scales. Further, to decrease the computational complexity, an image-level CNN is introduced to replace patch-level CNN in C2RNet. The proposed detection method learns the differences of various image properties to guarantee a stable detection performance, and the image-level CNN tremendously decreases its computational time. After the suspicious forgery regions are located by the proposed C2RNet, the final detected forgery regions are generated by applying the proposed adaptive clustering approach. The experiment results demonstrate that the proposed detection method achieves relatively promising results compared with state-of-the-art splicing forgery detection methods, even under various attack conditions.
       
  • General Decay Lag Anti-synchronization of Multi-weighted Delayed Coupled
           Neural Networks with Reaction–Diffusion Terms
    • Abstract: Publication date: Available online 23 September 2019Source: Information SciencesAuthor(s): Yanli Huang, Jie Hou, Erfu Yang We propose a new anti-synchronization concept, called general decay lag anti-synchronization, by combining the definitions of decay synchronization and lag synchronization. Novel criteria for the decay lag anti-synchronization of multi-weighted delayed coupled reaction–diffusion neural networks (MWDCRDNNs) with and without bounded distributed delays are derived by constructing an appropriate nonlinear controller and using the Lyapunov functional method. Moreover, the robust decay lag anti-synchronization of MWDCRDNNs with and without bounded distributed delays is considered. Finally, two numerical simulations are performed to validate the obtained results.
       
  • Offset-Free State-Space Nonlinear Predictive Control for Wiener Systems
    • Abstract: Publication date: Available online 23 September 2019Source: Information SciencesAuthor(s): Maciej Ławryńczuk, Piotr Tatjewski This work is concerned with state space Multiple-Input Multiple-Output (MIMO) Wiener systems which consist of a linear dynamic block connected in series with a nonlinear steady-state (static) one. Model Predictive Control (MPC) algorithms with successive on-line model or trajectory linearisation for dynamic processes described by such Wiener systems are discussed. Advantages of the presented MPC algorithms are: a) computational efficiency since quadratic optimisation problems are only solved on-line, nonlinear optimisation is not necessary, b) very good quality of control, c) offset-free control (no steady-state error in presence of disturbances) assured by a novel approach to disturbance modelling and state estimation, resulting in a simple design and a simple control structure. All features of the discussed algorithms are demonstrated and their performance is compared with that of the MPC algorithm with nonlinear optimisation as well as with the traditional offset-free state-space MPC approach.
       
  • Novel results on synchronization for a class of switched inertial neural
           networks with distributed delays
    • Abstract: Publication date: Available online 23 September 2019Source: Information SciencesAuthor(s): Guodong Zhang, Zhigang Zeng, Di Ning This paper presents a class of state-dependent switched neural networks with inertial items and distributed delays. Several new results are derived to ensure the exponential synchronization of such switched neural networks by using a novel hybrid control scheme and the Lyapunov stability theory. Finally, simulations are given to show the validity of the derived results. We believe the new useful control method of this paper widens the application scope for the switched neural networks.
       
  • Interval-valued hesitant fuzzy multi-granularity three-way decisions in
           consensus processes with applications to multi-attribute group decision
           making
    • Abstract: Publication date: Available online 22 September 2019Source: Information SciencesAuthor(s): Chao Zhang, Deyu Li, Jiye Liang Multi-attribute group decision making (MAGDM) is a common activity for multi-variable complicated decision making situations by integrating collective wisdom. Aiming at fusing granular computing with three-way decisions (3WD) to study scheme synthesis and analysis of solution space, multi-granularity three-way decisions (MG-3WD) provide multi-dimension problem solving methods for MAGDM problems. By using MG-3WD frameworks, this paper intends to study viable strategies of processing consensus and conflicting opinions provided by different decision makers in the interval-valued hesitant fuzzy (IVHF) MAGDM problem. More specifically, after reviewing the relevant literature, four kinds of IVHF multigranulation decision-theoretic rough sets (MG-DTRSs) over two universes are proposed according to different risk appetites of experts firstly. Then, we explore some fundamental propositions of newly proposed models. Afterwards, solutions to MAGDM problems in the context of mergers and acquisitions (M&A) target selections by using the presented IVHF MG-DTRSs over two universes are constructed. At last, a M&A target selection case study, along with a sensitivity analysis and a comparative analysis, is applied to illustrate the established decision making approaches.
       
  • BotMark: Automated botnet detection with hybrid analysis of flow-based and
           graph-based traffic behaviors
    • Abstract: Publication date: Available online 20 September 2019Source: Information SciencesAuthor(s): Wei Wang, Yaoyao Shang, Yongzhong He, Yidong Li, Jiqiang Liu The Botnets have become one of the most serious threats to cyber infrastructure. Most existing work on detecting botnets is based on flow-based traffic analysis by mining their communication patterns. There also exists related work based on anomaly detection in communication graphs. As bots have continuously evolved and become increasingly sophisticated, only using flow-based traffic analysis or graph-based analysis for the detection would result in false negatives or false positives, or can even be evaded. In this work, we propose BotMark, an automated model that detects botnets with hybrid analysis of flow-based and graph-based network traffic behaviors. We extract 15 statistical flow-based traffic features as well as 3 graph-based features in building the detection model. For flow-based detection, we consider the similarity and stability of C-flow as measurements in the detection. In particular, we employ k-means to measure the similarity of C-flows and assign similarity scores, and calculate stability score of C-flows through the distribution of packet length within a C-flow. The graph-based detection is based on the observation that the neighborhoods of anomalous nodes significantly differ from those of normal nodes in communication graphs. In particular, we use least-square technique and Local Outlier Factor (LOF) to calculate anomaly scores that measure the differences of their neighborhoods. Our models use the scores to mark bots. BotMark performs automated botnet detection with hybrid analysis of flow-based and graph-based traffic behaviors by ensemble of the detection results based on similarity scores, stability scores and anomaly scores. We collect a very large size of network traffic by simulating 5 newly propagated botnets, including Mirai, Black energy, Zeus, Athena and Ares in a real computing environment. Extensive experimental results demonstrate the effectiveness of BotMark. It achieves 99.94% in terms of detection accuracy, outperforming any individual detector with flow-based detection or graph-based detection.
       
  • On the f-divergence for discrete non-additive measures
    • Abstract: Publication date: Available online 19 September 2019Source: Information SciencesAuthor(s): Vicenç Torra, Yasuo Narukawa, Michio Sugeno In this paper we study the definition of the f-divergence and the Hellinger distance for non-additive measures in the discrete case. As these measures are based on the derivatives of the measures, we consider the problem of defining the Radon-Nikodym derivative of a non-additive measure.While Radon-Nikodym derivatives for additive measures exist for absolutely continuous measures, this is not the case in the non-additive case. In this paper we will define set-directional and upper, lower and interval derivatives. We will also define when two measures have the same sign. These definitions will be used to introduce alternative definitions of the f-divergence, all extending the classical definition to non-additive measures.
       
  • Adaptive robust control of oxygen excess ratio for PEMFC system based on
           type-2 fuzzy logic system
    • Abstract: Publication date: Available online 12 September 2019Source: Information SciencesAuthor(s): H.K. Zhang, Y.F. Wang, D.H. Wang, Y.L. Wang The Proton Exchange Membrane Fuel Cell (PEMFC) air supply system takes on the characteristics of external disturbances and uncertain parameters, which is difficult to achieve accurate modeling and stability control. In this paper, an adaptive robust controller based on type-2 fuzzy logic systems (T2-FLS) is proposed to control the oxygen excess ratio (OER) of PEMFC air supply system. The controller does not need the unmodeled dynamics, which can be approximated by adaptive T2-FLS whose adaptive parameters are derived based on Lyapunov theory. The stability analysis shows that the system tracking error is uniform ultimate bounded. Finally, the practicability and feasibility of controller are validated by numerical simulation and Hardware-In-Loop (HIL) experiment.
       
 
 
JournalTOCs
School of Mathematical and Computer Sciences
Heriot-Watt University
Edinburgh, EH14 4AS, UK
Email: journaltocs@hw.ac.uk
Tel: +00 44 (0)131 4513762
Fax: +00 44 (0)131 4513327
 
Home (Search)
Subjects A-Z
Publishers A-Z
Customise
APIs
Your IP address: 3.231.226.211
 
About JournalTOCs
API
Help
News (blog, publications)
JournalTOCs on Twitter   JournalTOCs on Facebook

JournalTOCs © 2009-