Search

Abstract: In this paper we introduce a notion of a t-filter on residuated lattices which is a generalization of several special types of filters. We provide some basic properties of t-filters and show how particular results about special types of filters (e.g. Extension property, Triple of equivalent characteristics, and Quotient characteristics) are uniformly covered by this simple general framework. [Copyright &y& Elsevier]

In recent decades, the development of social computing systems has realized the efficient information exchange between large groups of people. Nowadays, social computing systems are rather complex platforms supported by not only traditional sociology theory but also computer science and big data based applications. With the increase of the social computing systems' complexities, serious issues of social digital security and privacy have shown up since, in recent years, more and more social data leakage incidents are happening. This fact is due to reasons on many different aspects since there are many sources threatening the security and privacy of the social data in such a complex social computing system. In this paper, we improve the traditional social data protection schemes by combining the information fragmentation concepts with the distributed system architectures to build a novel social data protection scheme. We use social photo protection as the fundamental scenario and deploy our novel scheme to illustrate the improvement on the protection level with the protection analysis in detail. A security analysis of practically realizing such a scheme is also evaluated in this paper. [ABSTRACT FROM AUTHOR]

In this work, with the help of a calculus system syllogistic logic with Carroll's diagrams (SLCD), we construct a useful algorithm for the possible deductions of polysyllogisms (soriteses). This algorithm makes a general deduction in categorical syllogisms with the help of diagrams to depict each proposition of polysyllogisms. The developed calculus system PolySLCD (PSLCD) is used to allow a formal deduction from premises set by comprising synchronically biliteral and triliteral diagrammatical appearance and simple algorithmic nature. This algorithm can be used to deduce new conclusions, step by step, through recursive conclusion sets that are obtained from premises of categorical polysyllogisms. The fundamental contributions of this paper are accurately deducing conclusions from sets corresponding to given premises as exact human reasoning using a single algorithm and designing this algorithm based on SLCD. Therefore, it is more suitable for computer-aided solution. Since the algorithm is set-based, it is a novel algorithm in the literature and it can easily contribute to the researchers using polysyllogisms in different scientific branches, such as computer science, decision-making systems and artificial intelligence. [ABSTRACT FROM AUTHOR]

Link Recommendation (LR) in complex networks has attracted huge interest in the social and computer science communities. Numerous networks, such as recommendation systems and social networks (which facilitate user contact), are probabilistic rather than deterministic due to the uncertainty surrounding the presence of links. Evaluating the various characteristics of such networks has frequently been tricky as the Intra-layer linkage graph requires at least two nodes to be in the same layer. Moreover, many existing LR methods mainly operate well on Single-Layer Graphs (SLGs) compared to Multi-Layer Graphs (MLGs) when nodes traverse multi-layers in a network of Intra-layer linkages. Considering this drawback, this paper proposes a Multi-Layer Stochastic Block Interaction method driven by Logistic Regression (MLSBI-LR) to exploit the bi-directional resources associated with Intra-layer linkages. Its inherent dependence on knowledge-based systems uses multi-criteria recommender systems to accommodate additional criteria and can modify neighborhood-based approaches. A multi-criteria network with relations over the same set of nodes is used since the modified neighbor-based method can exhibit rich dependence between entities and have available experimental data sets in MLGs to recommend links that would efficiently enrich users' experience. The accuracy and robustness of the proposed MLSBI-LR method compared to existing LR methods were extensively investigated using three distinct benchmark data sets and four evaluation metrics. Based on our experimental results across the databases and metrics, the proposed MLSBI-LR method performed significantly better (recording up to 17% increment in accuracy), recommending potential links in MLGs. Consequently, the proposed method may revolutionize link recommendation tasks in social networks by improving users' overall experience. • Multi-Layer Stochastic Block Interaction driven by Logistic Regression Proposed. • Recommendation uses bi-directional resources associated with Intra-layer linkages. • Uncertainty graph and machine learning enhances the Intra-layer linkage graphs. • Structural properties reveal important information about interaction in the system. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

In this paper, we present a new importance sampling method, namely the conditional importance sampling (CIS). This new method uses a conditional density as a proposal density and exploits rejection sampling, adaptively neglecting samples whose importance weights are relatively low. The CIS improves the efficiency of estimation without creating bias. We apply the CIS to the bootstrap filter to obtain a new algorithm, named the conditional bootstrap filter, which achieves higher estimation efficiency than the bootstrap filter and shows advantages over some other filters in our simulations. [ABSTRACT FROM AUTHOR]

Multi-modal optimization is an active research topic that has attracted increasing attention from evolutionary computation community. Particle swarm optimization (PSO) with niching technique is one of the most effective approaches for multi-modal optimization. However, in existing PSO with niching methods, the number of particles around different niches varies distinctly from each other, which makes it difficult for the algorithm to find high-quality solutions in all niches. To address this issue, this paper proposes a new niching PSO with equilibrium factor named E-SPSO. Different from the existing niching PSOs, the numbers of particles in different niches have been kept in balance in E-SPSO. The velocity of each particle is influenced by not only the personal best particle and the global best particle, but also an equilibrium factor (EF). By using the equilibrium factor to update the velocities of particles, the particles can be allocated uniformly among the niches. In this way, the computation resources can be assigned to the niches in a more balanced manner, so that the algorithm can gain more population diversity and find high-quality solutions in all niches. Experimental results on eleven benchmark problems show that the proposed mechanism not only increases the number of optima found, but also improves the search efficiency. [ABSTRACT FROM AUTHOR]

Target-level sentiment analysis (TLSA) is a classification task to extract sentiments from targets in text. In this paper, we propose t arget-dependent c onvolutional n eural n etwork (TCNN) tailored to the task of TLSA. The TCNN leverages the distance information between the target word and its neighboring words to learn the importance of each word to the target. Experimental results show that the TCNN achieves state-of-the-art performance on both single- and multi-target datasets. Qualitative evaluations were conducted to demonstrate the limitations of previous TLSA methods and also to verify that distance information is crucial for TLSA. Furthermore, by exploiting a convolutional neural network (CNN), the TCNN trains six times faster per epoch than other baselines based on recurrent neural networks. [ABSTRACT FROM AUTHOR]

This article explores the problem of formulating a general measure of sparsity of network graphs. Based on an available definition sparsity of a dataset, namely Gini index, it provides a way to define sparsity measure of a network graph. We name the sparsity measure so introduced as sparsity index . Sparsity measures are commonly associated with six properties, namely, Robin Hood, Scaling, Rising Tide, Cloning, Bill Gates and Babies. Sparsity index directly satisfies four of these six properties; does not satisfy Cloning and satisfies Scaling for some specific cases. A comparison of the proposed index is drawn with Edge Density (the proportion of the sum of degrees of all nodes in a graph compared to the total possible degrees in the corresponding fully connected graph), by showing mathematically that as the edge density of an undirected graph increases, its sparsity index decreases. The paper highlights how the proposed sparsity measure can reveal important properties of a network graph. Further, a relationship has been drawn analytically between the sparsity index and the exponent term of a power law distribution (a distribution known to approximate the degree distribution of a wide variety of network graphs). To illustrate application of the proposed index, a community detection algorithm for network graphs is presented. The algorithm produces overlapping communities with no input requirement on number or size of the communities; has a computational complexity O ( n 2 ), where n is the number of nodes of the graph. The results validated on artificial and real networks show its effectiveness. [ABSTRACT FROM AUTHOR]

Abstract: In a recent paper Greco et al. (2012) propose a number of properties for measures of rule interestingness. The most fundamental of these properties is that such measures should be Bayesian confirmation measures and this criterion provides the context for the current paper as well. They also propose a number of properties relating to entailment and symmetry in order to discriminate between various confirmation measures which have been proposed in the literature. Working within the same framework of confirmation measures, several limitations of their proposed properties are discussed and a motivation provided for alternative properties. Two new measures of interestingness are proposed and then compared with two other recently proposed measures which also satisfy these properties. [Copyright &y& Elsevier]

Information Sciences is a leading international journal in computer science launched in 1968, so becoming fifty years old in 2018. In order to celebrate its anniversary, this study presents a bibliometric overview of the leading publication and citation trends occurring in the journal. The aim of the work is to identify the most relevant authors, institutions, countries, and analyze their evolution through time. The paper uses the Web of Science Core Collection in order to search for the bibliographic information. Our study also develops a graphical mapping of the bibliometric material by using the visualization of similarities (VOS) viewer. With this software, the work analyzes bibliographic coupling, citation and co-citation analysis, co-authorship, and co-occurrence of keywords. The results underline the significant growth of the journal through time and its international diversity having publications from countries all over the world. [ABSTRACT FROM AUTHOR]

Distributed networked control systems have attracted intense attention from both academia and industry due to the multidisciplinary nature among the areas of communication networks, computer science and control. With ever-increasing research trends in these areas, it is desirable to review recent advances and to identify methodologies for distributed networked control systems. This paper presents a brief overview of such systems regarding system configurations, challenging issues and methodologies. First, networked control systems are introduced and their prevalent configurations including centralized, decentralized and distributed structures are outlined. Second, an emphasis is laid on a number of challenging issues from the analysis and synthesis of distributed networked control systems. More specifically, these challenging issues are identified through three integrated aspects: communication, computation and control. Third, different methodologies in the literature for distributed networked control systems are reviewed and categorized based on three pairs: undirected and directed graphs, fixed and time-varying topologies, and time-triggered and event-triggered mechanisms. Finally, concluding remarks are drawn and some potential research directions are suggested. [ABSTRACT FROM AUTHOR]

Multilevel decision-making techniques aim to deal with decentralized management problems that feature interactive decision entities distributed throughout a multiple level hierarchy. Significant efforts have been devoted to understanding the fundamental concepts and developing diverse solution algorithms associated with multilevel decision-making by researchers in areas of both mathematics/computer science and business areas. Researchers have emphasized the importance of developing a range of multilevel decision-making techniques to handle a wide variety of management and optimization problems in real-world applications, and have successfully gained experience in this area. It is thus vital that a high quality, instructive review of current trends should be conducted, not only of the theoretical research results but also the practical developments in multilevel decision-making in business. This paper systematically reviews up-to-date multilevel decision-making techniques and clusters related technique developments into four main categories: bi-level decision-making (including multi-objective and multi-follower situations), tri-level decision-making, fuzzy multilevel decision-making, and the applications of these techniques in different domains. By providing state-of-the-art knowledge, this survey will directly support researchers and practical professionals in their understanding of developments in theoretical research results and applications in relation to multilevel decision-making techniques. [ABSTRACT FROM AUTHOR]

In recent years, several mathematical concepts have been successfully explored in the computer science domain as a basis for finding original solutions for complex problems related to knowledge engineering, data mining, and information retrieval. Hence, relational algebra (RA) and formal concept analysis (FCA) may be considered as useful mathematical foundations that unify data and knowledge into information retrieval systems. For example, some elements in a fringe relation (related to the (RA) domain) called isolated points have been successfully used in FCA as formal concept labels or composite labels. Once associated with words in a textual document, these labels constitute relevant features of a text. This paper proposes the MinGenCoverage algorithm for covering a Formal Context (as a formal representation of a text) based on isolated labels and using these labels (or text features) for categorization, corpus structuring, and micro–macro browsing as an advanced information retrieval functionality. The main thrust of the approach introduced here relies heavily on the close connection between isolated points and minimal generators (MGs). MGs stand at the antipodes of the closures within their respective equivalence classes. By using the fact that the minimal generators are the smallest elements within an equivalence class, their detection and traversal is greatly eased and the coverage can be swiftly built. Extensive experiments provide empirical evidence for the performance of the proposed approach. [ABSTRACT FROM AUTHOR]

The problem of link prediction has recently attracted considerable attention in various domains, such as sociology, anthropology, information science, and computer science. In many real world applications, we must predict similarity scores only between pairs of vertices in which users are interested, rather than predicting the scores of all pairs of vertices in the network. In this paper, we propose a fast similarity-based method to predict links related to nodes of interest. In the method, we first construct a sub-graph centered at the node of interest. By choosing the proper size for such a sub-graph, we can restrict the error of the estimated similarities within a given threshold. Because the similarity score is computed within a small sub-graph, the algorithm can greatly reduce computation time. The method is also extended to predict potential links in the whole network to achieve high process speed and accuracy. Experimental results on real networks demonstrate that our algorithm can obtain high accuracy results in less time than other methods can. [ABSTRACT FROM AUTHOR]

In this paper we propose and solve common influence region problems. These problems are related to the simultaneous influence, or the capacity to attract customers, of two sets of facilities of different types. For instance, while a facility of the first type competes with the other facilities of the first type, it cooperates with several facilities of the second type. The problems studied can be applied, for example, to decision-making support systems for marketing and/or locating facilities. We present parallel algorithms, to be run on a Graphics Processing Unit, for approximately solving the problems considered here. We also provide experimental results and discuss the efficiency and scalability of our approach. Finally, we present the speedup ratios obtained when the running times of the parallel proposed algorithms using a GPU are compared with those obtained from their respective efficient sequential CPU versions. [ABSTRACT FROM AUTHOR]

This paper defines a temporal logic for reaction systems (rsCTL). The logic is interpreted over the models for the context restricted reaction systems that generalise standard reaction systems by controlling context sequences. Moreover, a translation from the context restricted reaction systems into boolean functions is defined in order to be used for a symbolic model checking for rsCTL over these systems. The model checking for rsCTL is proved to be pspace -complete. The proposed approach to model checking was implemented and experimentally evaluated using four benchmarks. [ABSTRACT FROM AUTHOR]

Learning a distance metric from training samples is often a crucial step in machine learning and pattern recognition. Locality, compactness and consistency are considered as the key principles in distance metric learning. However, the existing metric learning methods just consider one or two of them. In this paper, we develop a multi-granularity distance learning technique. First, a new index, neighborhood granule margin, which simultaneously considers locality, compactness and consistency of neighborhood, is introduced to evaluate a distance metric. By maximizing neighborhood granule margin, we formulate the distance metric learning problem as a sample pair classification problem, which can be solved by standard support vector machine solvers. Then a set of distance metrics are learned in different granular spaces. The weights of the granular spaces are learned through optimizing the margin distribution. Finally, the decisions from different granular spaces are combined with weighted voting. Experiments on UCI datasets, gender classification and object categorization tasks show that the proposed method is superior to the state-of-the-art distance metric learning algorithms. [ABSTRACT FROM AUTHOR]

Due to the critical role of workflows in organizations, their design must be assisted by automatic formal verification approaches. The aim is to prove formally, before implementation, their correctness w.r.t. the required properties such as achieving safely the expected services (soundness property). In this perspective, time workflow nets (TWF-nets for short) are proposed as a framework to specify and verify the soundness of workflows. The verification process is based on state space abstractions and takes into account the time constraints of workflows. However, it suffers from the state explosion problem due the interleaving semantics of TWF-nets. To attenuate this problem, this paper investigates the combination of a state space abstraction with a partial order reduction technique. Firstly, it shows that to verify soundness of a TWF-net, it suffices to explore its non-equivalent firing sequences. Then, it establishes a selection procedure of the subset of transitions to explore from each abstract state and proves that it covers all its non-equivalent firing sequences. Finally, the effectiveness of the proposed approach is assessed by some experimental results. [ABSTRACT FROM AUTHOR]

Microarray data classification is a difficult challenge for machine learning researchers due to its high number of features and the small sample sizes. Feature selection has been soon considered a de facto standard in this field since its introduction, and a huge number of feature selection methods were utilized trying to reduce the input dimensionality while improving the classification performance. This paper is devoted to reviewing the most up-to-date feature selection methods developed in this field and the microarray databases most frequently used in the literature. We also make the interested reader aware of the problematic of data characteristics in this domain, such as the imbalance of the data, their complexity, or the so-called dataset shift. Finally, an experimental evaluation on the most representative datasets using well-known feature selection methods is presented, bearing in mind that the aim is not to provide the best feature selection method, but to facilitate their comparative study by the research community. [ABSTRACT FROM AUTHOR]

Facing the massive amount of data appearing on the web, automatic analysis tools have become essential for web users to discover valuable information online. Precise similarity measurement plays a decisive role in enabling analysis tools to acquire high-quality performances. Because different features contribute diversely to similarity calculation, it is necessary to utilize weight to measure feature's contribution and import it into similarity measurement. To accurately assign feature's weight, constrained data-pairs provided by users are usually imported into the weight evaluation procedure, whereas conventional plans all fail to consider two challenges: (a) asymmetrical distribution of constrained data-pairs, and (b) inconsistency contained by constrained data-pairs. If these two issues occur, conventional plans are incompetent at addressing them or are even unable to work. Thus, this paper proposes a novel constraint based weight evaluation to address these two issues. For the former, constrained data-pairs are partitioned into several equivalent classes, and distributing parameters are assigned to constrained data-pairs to balance their distributions. For the latter, constrained data-pairs are connected one after another, and belief values are thereby formed to indicate their probability of being inconsistent. Experimental results demonstrate that this type of evaluation is independent of any algorithm. With this evaluation, similarities can be calculated more accurately. [ABSTRACT FROM AUTHOR]

Abstract: The limit inferior and limit superior of a bounded sequence of fuzzy real numbers have been introduced by Aytar et al. (2008) [1]. In this paper we give a simplified expressions for limit inferior and limit superior. The expressions is concise and convenient for use. As a straightforward corollary of this expressions, we can easily prove some properties of the limit inferior and limit superior. [Copyright &y& Elsevier]

Abstract: Evolutionary membrane computing is an important research direction of membrane computing that aims to explore the complex interactions between membrane computing and evolutionary computation. These disciplines are receiving increasing attention. In this paper, an overview of the evolutionary membrane computing state-of-the-art and new results on two established topics in well defined scopes (membrane-inspired evolutionary algorithms and automated design of membrane computing models) are presented. We survey their theoretical developments and applications, sketch the differences between them, and compare the advantages and limitations. [Copyright &y& Elsevier]

Abstract: It is already known that large complete oriented stars (COSs) with intrinsic weights are amplifiers of selection. This paper addresses the evolutionary dynamics on COSs. First, we give the exact formulas for the fixation probabilities on a COS. We then apply these formulas to study some properties of COSs. The obtained results partially reveal the way in which the fixation probabilities are affected by COSs. [Copyright &y& Elsevier]

Abstract: Hierarchical clustering problem is a traditional topic in computer science, which aims to discover a consistent hierarchy of clusters with different granularities. One of the most important open questions on hierarchical clustering is the identification of the meaningful clustering levels in the hierarchical structure. In this paper, we answer this question from algorithmic point of view. In particular, we derive a quantitative analysis on the impact of the low-level clustering costs on high level clusters, when agglomerative algorithms are run to construct the hierarchy. This analysis enables us to find meaningful clustering levels, which are independent of the clusters hierarchically beneath it. We thus propose a general agglomerative hierarchical clustering framework, which automatically constructs meaningful clustering levels. This framework is proven to be generally applicable to any k-clustering problem in any -relaxed metric space, in which strict triangle inequality is relaxed within some constant factor . To fully utilize the hierarchical clustering framework, we conduct some case studies on k-median and k-means clustering problems, in both of which our framework achieves better approximation factor than the state-of-the-art methods. We also extend our framework to handle the data stream clustering problem, which allows only one scan on the whole data set. By incorporating our framework into Guha’s data stream clustering algorithm, the clustering quality is greatly enhanced with only small extra computation cost incurred. The extensive experiments show that our proposal is superior to the distance based agglomerative hierarchical clustering and data stream clustering algorithms on a variety of data sets. [Copyright &y& Elsevier]

Abstract: In this paper we study a problem that arises in the competitive facility location field. Facilities and customers are represented by points of a planar Euclidean domain. We associate a weighted distance to each facility to reflect that customers select facilities depending on distance and importance. We define, by considering weighted distances, the k-influence region of a facility as the set of points of the domain that has the given facility among their k-nearest/farthest neighbors. On the other hand, we partition the domain into subregions so that each subregion has a non-negative weight associated to it which measures a characteristic related to the area of the subregion. Given a weighted partition of the domain, the k-influence region problem finds the points of the domain where are new facility should be opened. This is done considering the known weight associated to the new facility and ensuring a minimum weighted area of its k-influence region. We present a GPU parallel approach, designed under CUDA architecture, for approximately solving the k-influence region problem. In addition, we describe how to visualize the solutions, which improves the understanding of the problem and reveals complicated structures that would be hard to capture otherwise. Integration of computation and visualization facilitates decision makers with an iterative what-if analysis process, to acquire more information to obtain an approximate optimal location. Finally, we provide and discuss experimental results showing the efficiency and scalability of our approach. [Copyright &y& Elsevier]