Search

Class imbalanced datasets are common across different domains including health, security, banking and others. A typical supervised learning algorithm tends to be biased towards the majority class when dealing with imbalanced datasets. The learning task becomes more challenging when there is also an overlap of instances from different classes. In this paper, we propose an undersampling framework for handling class imbalance in binary datasets by removing potential overlapped data points. Our methods are designed to identify and eliminate majority class instances from the overlapping region. Accurate identification and elimination of these instances maximise the visibility of the minority class instances and at the same time minimises excessive elimination of data, which reduces information loss. Four methods based on neighbourhood searching with different criteria to identify potential overlapped instances are proposed in this paper. Extensive experiments using simulated and real-world datasets were carried out. Results show comparable performance with state-of-the-art methods across different common metrics with exceptional and statistically significant improvements in sensitivity. [ABSTRACT FROM AUTHOR]

The traditional projection models in the bipartite networks involve many node pairs that consist of weak relationships. These node pairs lead to poor quality predictions as well as high computation time. In this paper, to cope with these problems, we firstly propose a novel projection model, which is called "Strengthened Projection Model". Then, to predict the potential links in the future, we present a new link prediction approach based on the proposed projection model. Thanks to the proposed model, the computation time is shortened, and the high probability predictions are extracted. The majority of the previous works conducted in this area have used the classical proximity measure algorithms that only take into account the current network structure, regardless of when events occur in the network evolution. To overcome these limited methods, in this paper, we also propose a novel proximity measure algorithm that considers the bipartite network evolution. To the best of our knowledge, this is the first attempt that takes into account the time-awareness in bipartite networks. To evaluate the performance of our proposed approach, we conducted experiments on the academic information network. To construct this bipartite network, we collected data from IEEE Xplore. The experimental results show that the success of the proposed method is promising. [ABSTRACT FROM AUTHOR]

• This paper is to find the truth Z of events in social sensing with information flows. • A truth discovery model is presented which considers the individuals' reliability R. • The model also considers the dependency D between individuals' observations. • An iterative expectation maximization is presented to infer both Z, R and D. • Extensive experiments demonstrate the effectiveness of the presented algorithm. Social sensing relies on a large number of observations reported by different, possibly unreliable, agents to determine if an event has occurred or not. In this paper, we consider the truth discovery problem in social sensing, in which an agent may receive another agent's observation (known as an information flow), and may change its observation to match the observation it receives. If an agent's observation is influenced by another agent, we say that the former is a dependent agent. We propose an Iterative Expectation Maximization algorithm for Truth Discovery (IEMTD) in social sensing with dependent agents. Compared with other popular truth discovery approaches, which assume either the agents' observations are independent, or their dependency is known a priori, IEMTD allows to infer each agent's reliability, the observations' dependency and the events' truth jointly. Simulation results on synthetic data and three real world data sets demonstrate that in almost all our experiments, IEMTD achieves a higher truth discovery accuracy than the existing algorithms when dependencies exist between agents' observations. [ABSTRACT FROM AUTHOR]

This paper is concerned with the problem of distributed Nash equilibrium computation in aggregative games. Note that the traditional computation algorithms are designed based on time-scheduled communication strategy, which may lead to high communication consumption of the whole network. To reduce the consumption, this paper proposes a novel distributed algorithm with an event-triggered mechanism, where the communication between any two agents is only carried out when an edge-based event condition is triggered. In the convergence analysis of the proposed algorithm, an important event-related error variable is firstly defined. Then, based on a zero-sum property of this event-related error, two key relations on the agents' estimates in the proposed algorithm are provided. Further, by using these relations, it is proven that the agents' estimates can achieve a Nash equilibrium under a proper event-triggering condition. Finally, examples on the demand response of power systems are presented to verify the theoretical findings. [ABSTRACT FROM AUTHOR]

Highlights • A new multi-method based bi-objective optimization algorithm is developed for solving bi-objective problems, with and without uncertainty. • An efficient technique is introduced to handle the uncertainty in uncertain bi-objective optimization problems. • A set of standard bi-objective and real-world problems, with and without uncertainity, are solved. • The performance of the algorithm is found to be better than other well-known approaches compared in this paper. Abstract Many real-world decision problems consider more than one objective. These objectives usually conflict with each other. In the problem environments, many data and parameters are usually uncertain at the time of planning. So multi-objective optimization combined with uncertainty is a challenging research topic. To deal with the basic search process in such problems, in this paper, a new approach has been proposed that combines multiple population based algorithms under a single algorithm structure. To deal with the uncertainty, a dynamic scenario-based approach is developed and integrated with the search process, in which the number of scenarios is dynamically set during the solution process. This is a new way of solving multi-objective optimization problems under uncertainty. To judge the performance of the proposed approach, we have solved a set of standard test problems without uncertainty and a number of practical problems with uncertainty. The practical problems are the well-known dynamic economic and emission dispatch problems, with different combinations of energy sources. The experimental studies demonstrated that the proposed approach performs better than other well-known algorithms compared in this paper. [ABSTRACT FROM AUTHOR]

Abstract Meta-learning has been successfully used for algorithm recommendation tasks. It uses machine learning to induce meta-models able to predict the best algorithms for a new dataset. In this paper, meta-models are applied to a set of meta-features, describing a dataset, to predict the performance of clustering algorithms applied to this dataset. The paper also proposes a new set of meta-features, based on correlation and dissimilarity measures. Experimental results show that these meta-features improve the recommendation. Additionally, this paper evaluates the importance of each meta-feature for the recommendation. [ABSTRACT FROM AUTHOR]

The Shortest Path in networks is an important problem incombinatorial optimization and has many applications in areas like telecommunications and transportation. It is known that this problem is easy to solve in its classic deterministic version, but it is also known that it is an NP-Hard problem for several generalizations. The Shortest Path Problem consists in finding a simple path connecting a source node and a terminal node in an arc-weighted directed network. In some real-world situations the weights are not completely known and then this problem is transformed into an optimization one under uncertainty. It is assumed that an interval estimate is given for each arc length and no further information about the statistical distribution of the weights is known. Uncertainty has been modeled in different ways in optimization. Our aim in this paper is to study the Minmax Regret path with interval data problem by presenting a new exact branch and cut algorithm and, additionally, new heuristics. A set of difficult and large size instances are defined and computational experiments are conducted for the analysis of the different approaches designed to solve the problem. The main contribution of our paper is to provide an assessment of the performance of the proposed algorithms and an empirical evidence of the superiority of a simulated annealing approach based on a new neighborhood over the other heuristics proposed. [ABSTRACT FROM AUTHOR]

In this paper we investigate an application of hybrid Monte Carlo Tree Search (MCTS) based algorithms to solving dynamic decision making problems. We employ UCT (the most popular MCTS approach) in combination with well-known Resource Constrained Project Scheduling Problem (RCPSP) and Stochastic Resource Constrained Project Scheduling Problem (SRCPSP) solvers to devise strategies for a generic and highly dynamic version of RCPSP, which we call Risk-Aware Project Scheduling Problem (RAPSP). We compare these strategies’ performance with results of both pure MCTS approach and non-MCTS solvers for projects of varied characteristics. We reach a conclusion that proposed hybrid simulation-heuristic methods are promising approaches to dynamic decision making problems, RAPSP in particular. Consequently, we argue that more research effort should be directed to applications of MCTS algorithm outside the domain of game-playing, with which it is commonly associated. At the same time, to the best of our knowledge, this paper is the first attempt at defining generalized SRCPSP model encompassing arbitrary risks and risk response / mitigation strategies as an optimization problem and applying Computational Intelligence methods to build fully-automated decision making systems. We strongly believe it to be a research direction worth further investigation, combining project scheduling, risk management and metaheuristic optimization techniques into a well-defined platform allowing direct comparisons of different strategies. [ABSTRACT FROM AUTHOR]

Source selection is the problem of identifying a subset of available data sources that best meet a user’s needs. In this paper we propose a user-driven approach to source selection that seeks to identify sources that are most fit for purpose. The approach employs a decision support methodology to take account of a user’s context, to allow end users to tune their preferences by specifying the relative importance between different criteria, looking to find a trade-off solution aligned with his/her preferences. The approach is extensible to incorporate diverse criteria, not drawn from a fixed set, and solutions can use a subset of the data from each selected source, rather than require that sources are used in their entirety or not at all. The paper describes and motivates the approach, presenting a methodology for modelling a user’s context, and its collection of optimisation algorithms for exploring the space of solutions, and compares and evaluates the resulting algorithms using multiple real world data sets. The experiments show how source selection results are produced that are attuned to each user’s preferences, both with respect to overall weighted utility and through faithful representation of a user’s preferences within a result, while scaling to potentially thousands of sources. [ABSTRACT FROM AUTHOR]

In order to address the problem of retinal image registration, this paper proposes and analyzes a novel and general matching algorithm called Multi-Attribute-Driven Regularized Mixture Model (MAD-RMM). Mismatches removal can play a key role in image registration, which refers to establish reliable matches between two point sets. Here the presented approach starts from multi-feature attributes which are used to guide the feature matching to identify inliers (correct matches) from outliers (incorrect matches), and then estimates the spatial transformation. In this paper, motivated by the problem of feature matching that the initial correspondence is always contaminated by outliers, thereby we formulate this issue as a probability deformable mixture model which consists of Gaussian components for inliers and uniform components for outliers. Moreover, the algorithm takes full advantage of using multiple attributes for better general matching performance. Here we are assuming all inliers are mapped into a high-dimensional feature space, namely reproducing kernel Hilbert space (RKHS), and the closed-form solution to the mapping function is given by the representation theorem with L 2 norm regularization under the Expectation-Maximization (EM) algorithm. Finally, we evaluate the performance of the algorithm by applying it to retinal image registration on several datasets, where experimental results demonstrate that the MAD-RMM outperforms current state-of-the-art methods and shows the robustness to outliers on real retinal images. [ABSTRACT FROM AUTHOR]

In constrained optimisation, the augmented Lagrangian method is considered as one of the most effective and efficient methods. This paper studies the behaviour of augmented Lagrangian function (ALF) in the solution space and then proposes an improved augmented Lagrangian method. We have shown that our proposed method can overcome some of the drawbacks of the conventional augmented Lagrangian method. With the improved augmented Lagrangian approach, this paper then proposes a cooperative coevolutionary differential evolution algorithm for solving constrained optimisation problems. The proposed algorithm is evaluated on a set of 24 well-known benchmark functions and five practical engineering problems. Experimental results demonstrate that the proposed algorithm outperforms the state-of-the-art algorithms with respect to solution quality as well as efficiency. [ABSTRACT FROM AUTHOR]

This paper considers parameter identification problems for multivariable Hammerstein controlled autoregressive moving average (CARMA) systems. We transform a multivariable Hammerstein CARMA system into a new identification model which contains a set of bilinear parameter vectors, i.e., two sets of parameter vectors. To solve the difficulty of estimating the two sets of parameter vectors, this paper constructs two different forms of the identification model in each of which the output is linear about one set of parameter vectors, and presents an extended stochastic gradient algorithm to interactively estimate the two sets of parameter vectors by using the hierarchical identification principle. The proposed hierarchical identification algorithm has a higher computational efficiency than the over-parametrization model based stochastic gradient algorithm. The simulation results indicate that the proposed algorithm is effective, and an example of modeling a photovoltaic array system with the multivariable Hammerstein model is given. [ABSTRACT FROM AUTHOR]

In general, heuristic optimization techniques lose some of the optimal solution of the objective function in the optimization process. This paper proposes a concept to retain those variables that might help in accelerating the complete optimization process. The motivation is to derive linkages between variables in a population set that will be used in crossover strategy. This crossover strategy is dependent on a deferred acceptance algorithm (DAA). Also, the property of linkages or interrelation is implemented to derive the relation between variables among dimensions. This paper proposes a linkage based deferred acceptance optimization (LDAO) technique. It is observed that the proposed algorithm has proved its efficacy on the set of unconstrained and constrained objective functions. Also, the proposed algorithm is tested on challenging real world problems (CEC 2011) and the functions present in CEC 2014 competition. [ABSTRACT FROM AUTHOR]

In this paper we present a new generalization of the mathematical notion of distance. It is based on the abstraction of the codomain of the distance function. The resulting functions must satisfy a generalized triangular inequality, which depends only on the order structure of the valuation space, i.e a monoid structure is not required. This type of functions will be called i-Distances (i-metrics, i-quasi-metric, etc.). We show that they generate a topology in a very natural way based on open balls. This paper generalizes (Santanaand Santiago, 2013) which has been successfully applied in the field of Clustering Algorithms (see Silva et al., 2014, 2015). An example in the field of Interval Mathematics is also investigated. The resulting topology is Hausdorf and regular but non-metrizable, what means that it cannot be generated by an usual metric. [ABSTRACT FROM AUTHOR]

Efficient computation of skyline probability over uncertain preferences has not received much attention in the database community as compared to skyline probability computation over uncertain data . All known algorithms for probabilistic skyline computation over uncertain preferences attempt to find inexact value of skyline probability by resorting to sampling or to approximation scheme. Exact computation of skyline probability for database with uncertain preferences of moderate size is not possible with any of the existing algorithms. In this paper, we propose an efficient algorithm that can compute skyline probability exactly for reasonably large database. The inclusion–exclusion principle is used to express skyline probability in terms of joint probabilities of all possible combination. In this regard we introduce the concept of zero-contributing set which has zero effect in the signed aggregate of joint probabilities. Our algorithm employs a prefix-based k-level absorption to identify zero-contributing sets. It is shown empirically that only a very small portion of exponential search space remains after level wise application of prefix-based absorption. Thus it becomes possible to compute skyline probability with respect to large datasets. Detailed experimental analysis for real and synthetic datasets are reported to corroborate this claim. We also propose an incremental algorithm to compute skyline probability in dynamic scenarios wherein objects are added incrementally. Moreover, the theoretical concepts developed in this paper help to devise an efficient technique to compute skyline probability of all objects in the database. We show that the exponential search space is pruned once and then for each individual object skyline probability can be derived by inspecting a portion of the pruned lattice. We also use a concept of revival of absorbed pairs . We believe that this process is more efficient than computing the skyline probability individually. [ABSTRACT FROM AUTHOR]

Given a set of weighted objects in a data space, the MaxRS problem in spatial databases studied in a VLDB 2012 paper is to find a location for a rectangular region of a given size such that the weighted sum of all the objects covered by the rectangular region centered at the optimal location is maximized. This problem is useful in lots of location-based service applications, such as finding the location for a new fast food restaurant with a limited delivery range attracting the greatest number of customers. The existing MaxRS problem assumes that the rectangular region is always placed horizontally and is non-rotatable. However, under this assumption, the weighted sum of all the covered objects may not be the greatest when the rectangular region is rotatable. In this paper, we propose a generalized MaxRS problem called rotating MaxRS without this assumption. In rotating MaxRS, the rectangular region is rotatable and can be associated with an inclination angle. The goal of our problem is to find a location and an inclination angle such that the weighted sum of all the objects covered by the rectangular region of a given size centered at this location with this inclination angle is the greatest. We also present an efficient algorithm for the problem. Extensive experiments were conducted to verify the efficiency of our algorithms based on the real and synthetic datasets. The experimental results show that the weighted sum of all the objects in the rotating MaxRS queries can be increased with up to 300% on the synthetic datasets compared with existing non-rotating MaxRS queries, which shows the significance of the new rotating MaxRS queries. [ABSTRACT FROM AUTHOR]

Bayesian learning provides a firm theoretical basis of the design and exploitation of algorithms in data-streams processing (preprocessing, change detection, hypothesis testing, clustering, etc.). Primarily, it relies on a recursive parameter estimation of a firmly bounded complexity. As a rule, it has to approximate the exact posterior probability density (pd), which comprises unreduced information about the estimated parameter. In the recursive treatment of the data stream, the latest approximate pd is usually updated using the treated parametric model and the newest data and then approximated. The fact that approximation errors may accumulate over time course is mostly neglected in the estimator design and, at most, checked ex post. The paper inspects the estimator design with respect to the error accumulation and concludes that a sort of forgetting (pd flattening) is an indispensable part of a reliable approximate recursive estimation. The conclusion results from a Bayesian problem formulation complemented by the minimum Kullback–Leibler divergence principle. Claims of the paper are supported by a straightforward analysis, by elaboration of the proposed estimator to widely applicable parametric models and illustrated numerically. [ABSTRACT FROM AUTHOR]

Abstract: The cloud computing data center has numerous hosts as well as application requests. It needs to allocate resources dynamically according to the varied demands of users. It should not only provide QoS such as short response time and high throughput but also achieve efficient power consumption. This paper first puts forward a reconfiguration framework based on a request prediction, which anticipates the application request volume in advance. To determine the objective of relatively optimal configuration, it can work out the allocation scheme which can improve the resource utilization ratio as well as lower energy consumption. In addition, a concept of Utility Ratio Matrix (URM) is put forward to represent allocations of hosts and Virtual Machines (VMs), and a reconfiguration algorithm based on request prediction is also presented. The algorithm will predict the application requests so as to work out the allocation scheme in advance. The algorithm can separate the reconfiguration computing from the real allocation so that it can avoid a time delay between the reconfiguration result and the varied demands, and can also reduce the energy consumption in data center. The corresponding analysis and experimental results indicate the feasibility of the reconfiguration algorithm in this paper. [Copyright &y& Elsevier]

Abstract: This paper focuses on credibilistic clustering approach. A data clustering method partitions unlabeled data sets into clusters and labels them for various goals such as computer vision and pattern recognition. There are different models for objective function-based fuzzy clustering such as Fuzzy C-Means (FCM), Possibilistic C-Mean (PCM) and their combinations. Credibilistic clustering is a new approach in this field. In this paper, a new credibilistic clustering model is introduced in which credibility measure is applied instead of possibility measure in possibilistic clustering. Also, in objective function, the separation of clusters is considered in addition to the compactness within clusters. The steps of clustering are designed based on this approach. Finally, the main issues about model are discussed, and the results of computational experiments are presented to show the efficiency of the proposed model. [Copyright &y& Elsevier]

Abstract: Join aggregation is an important operation in database systems to return aggregate information on the join of two or several tables. Compared with exact query, it is a better choice in many cases to return approximate result satisfying a user-specified confidence interval in a much faster response time. It is found that none of previous works can efficiently process approximate join aggregation on massive data with arbitrary accuracy. This paper proposes a novel algorithm -AJA ( -Approximate Join Aggregation) to obtain approximate join aggregate result with arbitrary confidence interval efficiently. Two data structures of low space overhead, JRS and JPIPT, are presented in this paper. -AJA first makes use of JRS to return a quick response. If the approximate result computed by JRS does not satisfy the given confidence interval, JPIPT is exploited to obtain enough random join tuples. This paper presents a novel sampling algorithm to acquire random JPIPT tuples of specified size and devises its correctness proof. A tuple fetching method is proposed to retrieve join tuples by the sampled JPIPT tuples in one-pass sequential scan on joined tables. The construction and maintenance algorithms of JPIPT and JRS are provided also in this paper. The experimental results show that -AJA obtains 3 times to 2 orders of magnitude speedup over the existing algorithms and runs 1 to 4 orders of magnitude faster than exact query. [Copyright &y& Elsevier]

Abstract: Generating Membership Functions (MFs) from data is one of the fundamental challenges associated with the applications of fuzzy set theory. This paper proposes a new two-step algorithm, which uses discretization methods for initial partitioning, to generate MFs from data. In the first step, discretization algorithm divides domain of attributes to several partitions, and then, in the second step, an MF is defined on each partition. Four different methods are proposed to define MFs in the second step: the first method is based on partition width, the second is based on standard deviation of examples, the third is based on Coverage Rate of Neighbor Partitions (NPCR) and the last one is based on Coverage Rate of Partition (PCR). Coverage rate of partition and coverage rate of neighbor partition are two new introduced parameters, which can be used for MF generation. In addition, this paper proposes a new MF generation algorithm, called Fuzzy Entropy Based Fuzzy Partitioning (FEBFP), which is a specific version of the proposed two-step algorithm with some modifications. FEBFP uses fuzzy entropy of partition to generate MFs and involves the parameters of MFs in the process of MF generation to combine two steps of the algorithm. Non-parametric statistical tests are used to compare Fuzzy Decision Trees (FDTs) induced using the MFs generated by the proposed methods (employing different discretization algorithms as well as four MF generation methods). Experimental results show that eight methods outperform the others in terms of both accuracy and number of nodes. Among them, trapezoidal MFs that are defined by PCR on partitions generated by Zeta discretization algorithm, outperform the others when the accuracy and complexity of FDT have the same degree of importance. Moreover, the results show that the PCR and NPCR MF definition methods perform better than the other ones. [Copyright &y& Elsevier]

Abstract: This paper presents a vectorized matrix parameters encoding aspect for an evolutionary computer vision approach to procedural tree modeling. A serialized fixed-size floating-point encoded tree parameter set consists of a set of auxiliary local and other global parameters. The main goal of paper is to lower problem dimensionality needed for encoding local parameters. For evolution simulation, differential evolution algorithm is used. The optimizer evolves a parameterized procedural model by fitting a set of its rendered images to a set of automatically preprocessed reference photo images. The reconstructed tree morphology is then used for reconstructed tree animation, to generate similar geometrical tree models based on similar morphology. Examples of reconstructed model animation are shown, such as simulation of its growth, sway in the wind, or adding leaves. [Copyright &y& Elsevier]

Abstract: This paper proposes a novel frequent itemsets mining algorithm called PIETM (Principle of Inclusion–Exclusion and Transaction Mapping). PIETM has three major features. First, similar to the mining process in Apriori, PIETM discovers frequent itemsets in a bottom-up manner. However, it reduces database scanning to only two times. Second, PIETM does not scan the database to count the itemsets’ support. Instead, it employs the Principle of Inclusion–Exclusion to calculate the support of candidate itemsets. Third, PIETM uses transaction intervals to map and store the transaction ids of each item, which facilitates the itemsets counting process. We also present experimental results comparing PIETM with existing algorithms in this paper. The results show that PIETM takes lower execution time than other methods when the dataset has numerous items. In summary, this paper makes three major contributions. First, it presents a new method that calculates the support of itemsets using a well-known set theory property. Second, it demonstrates the correctness of counting itemsets in PIETM. Third, our method demonstrates its suitability for different high-performance applications by combining the advantages of the simplicity of Apriori and efficiency of FP-growth. [Copyright &y& Elsevier]

Resource failures may happen from time to time in an automated manufacturing system (AMS) in production practice, leading to that most of deadlock control methods in the literature are not applicable. For a generalized system of simple sequential process with resources (GS3PR), this paper develops a robust deadlock control strategy when there exists a type of unreliable resources. To do so, after computing the system's elementary and dependent strict minimal siphons (SMSs), by using the concept of max′-controllability of siphons, we then check whether an elementary SMS is self-max′-controlled or not, and whether it contains unreliable resources. Afterwards, a constraint set for a siphon is introduced and a monitor is designed for each non-max′-controlled elementary SMS and self-max′-controlled elementary one that contains unreliable resources. Then, if a dependent SMS is max′-controlled with respect to the control depth variables of its elementary siphons, it needs no monitor; otherwise, we add a monitor for such a dependent SMS. Finally, a robust deadlock control algorithm is developed to keep each SMS to be max′-controlled, even if there exists a type of unreliable resources. The proposed method is demonstrated by using examples. [ABSTRACT FROM AUTHOR]

The Markov blanket (MB), a key concept in a Bayesian network (BN), is essential for large-scale BN structure learning and optimal feature selection. Many MB discovery algorithms that are either efficient or effective have been proposed for addressing high-dimensional data. In this paper, we propose a new algorithm for E fficient and E ffective MB discovery, called EEMB. Specifically, given a target feature, the EEMB algorithm discovers the PC (i.e., parents and children) and spouses of the target simultaneously and can distinguish PC from spouses during MB discovery. We compare EEMB with the state-of-the-art MB discovery algorithms using a series of benchmark BNs and real-world datasets. The experiments demonstrate that EEMB is competitive with the fastest MB discovery algorithm in terms of computational efficiency and achieves almost the same MB discovery accuracy as the most accurate of the compared algorithms. [ABSTRACT FROM AUTHOR]

• First, we introduce a general communication mode, that is switching topologies with a transmission delay, instead of fixed topology without any delays in the existing literature. • Second, for the sampled-data scheme, we assume that the sampling intervals are aperiodic and time-varying, which may be more practical in real-life applications than the periodic sampling. And an algorithm utilized to determine the maximal allowable upper bound of sampling intervals is also provided. • Third, we concern both the exogenous disturbances in multi-agent systems and the controller gain perturbations in consensus protocol. In this regard, the robust non-fragile H ∞ control approach is employed to realize the tracking consistency in the network. In this paper, the sampled-data non-fragile H ∞ consensus tracking problem for Lipschitz nonlinear multi-agent systems with switching topologies and exogenous disturbances is investigated. Each possible interaction topology in the switching topologies set is assumed to contain a directed spanning tree. With introducing a sampled-data mechanism, the information is only capable of being aperiodically transmitted in the network at each sampling instant and unavoidably subject to a transmission delay. Then a protocol collecting the delayed sampled-data information from neighboring agents is proposed not only to provide robustness against some level of controller gain perturbations, but also to regulate consensus performance with an H ∞ disturbance attenuation level. By using tools from algebraic graph theory and Lyapunov–Krasovskii functional technique, it is proved that the concerned consensus tracking problem is solvable if the resultant consensus error system can be asymptotically stabilized. Simulation results verify the theoretical analysis. [ABSTRACT FROM AUTHOR]

Fuzzy linguistic preference relations (FLPRs) are an efficient way to express qualitative judgments of decision makers (DMs). This paper proposes a new group decision making (GDM) method based on the multiplicative consistency and consensus of FLPRs. First, we propose the concepts of consistency index and acceptable multiplicative consistent FLPRs. Then, we propose a method to improve the consistency of a FLPR, by which an acceptable multiplicative consistent FLPR is derived. Subsequently, we propose the consensus index for measuring the agreement among DMs. With respect to some FLPRs have an unacceptable multiplicative consistency and an unacceptable consensus, we propose an integer programming model to make both their consistency and consensus better to yield improved FLPRs with acceptable multiplicative consistency and consensus. Furthermore, the DMs' comprehensive weight vector is determined. Then, we propose a new GDM algorithm following the multiplicative consistency and consensus of FLPRs. Finally, the feasibility and the applicability of the proposed GDM method are illustrated via an application example and some comparative analyses with the existing GDM methods. [ABSTRACT FROM AUTHOR]

• The performance of three crossover operators of NSGA-III algorithm on big data optimization problems are investigated. • By introducing the concept of Stud, several improved crossover operators of SBX, UC, and SI have been designed. • Ten variants of NSGA-III algorithms are proposed. • The performance of ten NSGA-III variants are further verified on big data optimization problems. Traditional multi-objective optimization evolutionary algorithms (MOEAs) do not usually meet the requirements for online data processing because of their high computational costs. This drawback has resulted in difficulties in the deployment of MOEAs for multi-objective, large-scale optimization problems. Among different evolutionary algorithms, non-dominated sorting genetic algorithm-the third version (NSGA-III) is a fairly new method capable of solving large-scale optimization problems with acceptable computational requirements. In this paper, the performance of three crossover operators of the NSGA-III algorithm is benchmarked using a large-scale optimization problem based on human electroencephalogram (EEG) signal processing. The studied operators are simulated binary (SBX), uniform crossover (UC), and single point (SI) crossovers. Furthermore, enhanced versions of the NSGA-III algorithm are proposed through introducing the concept of Stud and designing several improved crossover operators of SBX, UC, and SI. The performance of the proposed NSGA-III variants is verified on six large-scale optimization problems. Experimental results indicate that the NSGA-III methods with UC and UC-Stud (UCS) outperform the other developed variants. [ABSTRACT FROM AUTHOR]

• A non-progressive diffusion model is proposed to describe the competitive propagation of two pieces of information or products. • The problem of minimizing rumor spread in social networks is studied and proved to be NP-hard. • A novel heuristic based on diffusion dynamics is proposed to solve the problem of minimizing rumor spread and compared with two adaptations of well-known ones. • A series of experiments are conducted on four different networks. Rumors can propagate at great speed through social networks and produce significant damages. In order to control rumor propagation, spreading correct information to counterbalance the effect of the rumor seems more appropriate than simply blocking rumors by censorship or network disruption. In this paper, a competitive diffusion model, namely Linear Threshold model with One Direction state Transition (LT1DT), is proposed for modeling competitive information propagation of two different types in the same network. The problem of minimizing rumor spread in social networks is explored and a novel heuristic based on diffusion dynamics is proposed to solve this problem under the LT1DT. Experimental analysis on four different networks shows that the novel heuristic outperforms pagerank centrality. By seeding correct information in the proximity of rumor seeds, the novel heuristic performs as well as the greedy approach in scale-free and small-world networks but runs three orders of magnitude faster than the greedy approach. [ABSTRACT FROM AUTHOR]

As a representative sequential pattern mining problem, counting the frequency of serial episodes from a streaming sequence has drawn continuous attention in academia due to its wide application in practice, e.g., telecommunication alarms, stock market, transaction logs, bioinformatics, etc. Although a number of serial episodes mining algorithms have been developed recently, most of them are neither stream-oriented, as they require processing the whole dataset multiple times, nor time-aware, as they fail to take into account the time constraint of serial episodes. In this paper, we propose two novel one-pass algorithms, ONCE and ONCE+, each of which can respectively compute two popular frequencies of given episodes satisfying predefined time-constraint as signals in a stream arrives one-after-another. ONCE is only used for non-overlapped frequency where the occurrences of a serial episode in sequence are not intersected. ONCE+ is designed for the distinct frequency where the occurrences of a serial episode do not share any event. Theoretical study proves that our algorithm can correctly mine the frequency of target time constraint serial episodes in a given stream. Experimental study over both real-world and synthetic datasets demonstrates that the proposed algorithm can work, with little time and space, in signal-intensive streams where millions of signals arrive within a single second. Moreover, the algorithm has been applied in a real stream processing system, where the efficacy and efficiency of this work are tested in practical applications. [ABSTRACT FROM AUTHOR]

The use of agents to circle a moving target and estimate its states with bearing measurements only is a kind of passive tracking technique, and has received wide attention in recent years. However, the performance of the existing estimation and control laws for moving agent are still not so satisfactory when the target position is not known beforehand. In this paper, an unbiased estimator to locate the position of the target and a controller to circumnavigate around the target at a prescribed circular distance are investigated for an unmanned aircraft vehicle (UAV). First, a rough target-state estimator and a rough two-stage saturation UAV circumnavigation strategy are introduced, and their convergence performance is analyzed. Second, a virtual intersecting localization algorithm (VILA) is proposed based on a rough estimation strategy, and then the accurate target-state estimation is obtained with a Kalman filter. Third, the UAV rough circumnavigating strategy is modified with the accurate target-state estimation. By numerical simulation, it is verified that the steady estimation error with VILA is asymptotically unbiased, and the estimation accuracy is much higher than that of the existing estimation methods. It is also proven that the control error by the proposed UAV circumnavigating strategy can asymptotically converge to zero by the modified controller, which is also superior to that for the existing circumnavigating controller. [ABSTRACT FROM AUTHOR]

Mobile edge clouds can offer delay-sensitive services to users by deploying storage and computing resources at the network edge. Considering resource-limited edge server, it is impossible to deploy all services on the edge clouds. Thus, many existing works have addressed the problem of arranging services on mobile edge clouds for better quality of services (QoS) to users. In terms of existing service placement strategies, the historical data of requesting services by users are collected to analyze. However, those historical data belong to users' sensitive information, without appropriate privacy preserving measures may hinder the implementation of traditional service arrangement strategies. Service placement with considering users' privacy and limited resources of mobile edge clouds, is an extremely urgent problem to be solved. In this paper, we propose a privacy-aware service placement (PSP) scheme to address the problem of service placement with privacy-awareness in the edge cloud system. The purpose of PSP mechanism is to protect users' privacy and provide better QoS to users when obtaining services from mobile edge clouds. Specifically, whether service placement on mobile edge clouds or not is modeled as a 0–1 problem. Then, a hybrid service placement algorithm is proposed that combines a centralized greedy algorithm and distributed federated learning. Compared with other optimization schemes, the simulation experiments show that PSP algorithm could not only protect users' privacy but also meet users' service demands through mobile edge clouds. [ABSTRACT FROM AUTHOR]

• We provide fundamental definitions to mathematically describe the multiple ellipse fitting problem. • We propose a framework rather than design an algorithm for the fitting problem. • We employ the circularly linked list to store and handle the contour point set. • We apply fixed-size sliding windows to the fitting problem. Multiple ellipse fitting is challenging and at the same time essential as it has a variety of applications in biology, chemistry, and nanotechnology. Accurate, effective, and reliable approach for the fitting problem has been always desirable. In this paper, we address a category of multiple ellipse fitting problem which fits densely connected contours. We propose a framework rather than design an algorithm for the problem. The framework streamlines five processes which include: sorting the contour points, doing ellipse fitting in sliding windows, detecting the context anomaly, performing clustering, and obtaining multiple ellipses through second ellipse fitting. The framework is evaluated in a real-world application of handprint identification and various synthetic datasets. Experimental results show that the framework can extract multiple ellipses from contours with satisfactory accuracy and efficiency. [ABSTRACT FROM AUTHOR]

Clustering is a technique that allows data to be organized into groups of similar objects. DBSCAN (Density-Based Spatial Clustering of Applications with Noise) constitutes a popular clustering algorithm that relies on a density-based notion of cluster and is designed to discover clusters of arbitrary shape. The computational complexity of DBSCAN is dominated by the calculation of the ϵ-neighborhood for every object in the dataset. Thus, the efficiency of DBSCAN can be improved in two different ways: (1) by reducing the overall number of ϵ-neighborhood queries (also known as region queries), or (2) by reducing the complexity of the nearest neighbor search conducted for each region query. This paper deals with the first issue by considering the most relevant region query strategies for DBSCAN, all of them characterized by inspecting the neighborhoods of only a subset of the objects in the dataset. We comparatively evaluate these region query strategies (or DBSCAN variants) in terms of clustering effectiveness and efficiency; additionally, a novel region query strategy is introduced in this work. The results show that some DBSCAN variants are only slightly inferior to DBSCAN in terms of effectiveness, while greatly improving its efficiency. Among these variants, the novel one outperforms the rest. [ABSTRACT FROM AUTHOR]

Rapid development in GPS tracking techniques produces a large number of spatio-temporal trajectory data. The analysis of these data provides us with a new opportunity to discover behavioural patterns. Spatio-temporal periodic pattern mining is finding temporal regularities for interesting places. Mining periodic patterns from spatio-temporal trajectories reveals useful and important information about people's regular and recurrent movements and behaviours. Existing periodic pattern mining algorithms suffer from two main drawbacks. They assume regularly sampled and evenly spaced trajectory data as input which is unlike real world data, traditional methods also fail to consider background aspatial information despite many applications requiring a semantic interpretation of movement behaviours. In this paper, we propose a new semantic periodic pattern mining algorithm from spatio-temporal trajectories that overcomes these two drawbacks from past studies. Experimental results with real world datasets demonstrate the efficiency and effectiveness of our proposed method. [ABSTRACT FROM AUTHOR]

• The mathematical properties of Yao's optimization model are discussed. • The existence and uniqueness of the optimal solutions are analyzed. • Four kinds of constrained three-way approximations of fuzzy sets are introduced. • A fast algorithm to obtain the optimal solutions under constrained notions is designed. • Experiments are studied with some typical fuzzy sets and some data from UCI repository. Three-way approximations of fuzzy sets aim at abstracting fuzzy sets into three pair-wise disjoint categories which facilitate semantic-oriented interpretations and reduce computing burden. Shadowed sets are a schema of three-way approximations of fuzzy sets which are formed based on a specific optimization mechanism. Among different principles guiding the construction of shadowed sets, the criterion of minimum distance offers a new insight within the framework of three-way decision theory. In this paper, the essential mathematical properties of the objective function used as a criterion to construct three-way approximations of fuzzy sets based on the principle of minimal distance, as well the characteristics of the optimal solutions, are analyzed. It is demonstrated that this optimization objective function is continuous but nonconvex with respect to the optimized variables. The nonconvex property makes the solution difficult and different approximate region partitions are obtainable even under the same optimization model. Therefore, further criteria are required to select final partition thresholds and make the construction process well-defined. To address this limitation, the notion of constrained three-way approximations of fuzzy sets is proposed from the perspective of minimal distance. Moreover, a constructive algorithm is provided to obtain the proposed constrained three-way approximations rather than using a direct enumeration method, and its performance is illustrated by considering some typical fuzzy sets along with some data from UCI repository. [ABSTRACT FROM AUTHOR]

Gene expression data classification is an important technology for cancer diagnosis in bioinformatics and has been widely researched. Due to the large number of genes and the small sample size in gene expression data, feature selection based on neighborhood rough sets is a key step for improving the performance of gene expression data classification. However, some quantitative measures of feature sets may be nonmonotonic in neighborhood rough sets, and many feature selection methods based on evaluation functions yield high cardinality and low predictive accuracy. Therefore, investigating effective and efficient heuristic reduction algorithms is necessary. In this paper, a novel feature selection method based on neighborhood rough sets using neighborhood entropy-based uncertainty measures for cancer classification from gene expression data is proposed. First, some neighborhood entropy-based uncertainty measures are investigated for handling the uncertainty and noise of neighborhood decision systems. Then, to fully reflect the decision-making ability of attributes, the neighborhood credibility and neighborhood coverage degrees are defined and introduced into decision neighborhood entropy and mutual information, which are proven to be nonmonotonic. Moreover, some of the properties and relationships among these measures are derived, which is helpful for understanding the essence of the knowledge content and the uncertainty of neighborhood decision systems. Finally, the Fisher score method is employed to preliminarily eliminate irrelevant genes to significantly reduce complexity, and a heuristic feature selection algorithm with low computational complexity is presented to improve the performance of cancer classification using gene expression data. Experiments on ten gene expression datasets show that our proposed algorithm is indeed efficient and outperforms other related methods in terms of the number of selected genes and the classification accuracy, especially as the size of the genes increases. [ABSTRACT FROM AUTHOR]

• We propose hypothetical distribution models with theoretical method to compute the optimal number of query labels. • We present clustering-based practical statistical method for the same issue. • We design the cost-sensitive active learning through statistical methods (CATS) algorithm. • Results of significance test verify the superiority of CATS to state-of-the-art algorithms. Clustering-based active learning splits the data into a number of blocks and queries the labels of most the representative instances. When the cost of labeling and misclassification are considered, we also face a key issue: How many labels should be queried for a given block. In this paper, we present theoretical and practical statistical methods to handle this issue. The theoretical statistical method calculates the optimal number of query labels for a predefined label distribution. Considering label distributions for different clustering qualities, we obtain three hypothetical models, namely Gaussian, Uniform, and V models. The practical statistical method calculates empirical label distribution of the cluster blocks. Considering four popular clustering algorithms, we use symmetry and curve fitting techniques on 30 datasets to obtain empirical distributions. Inspired by three-way decision, we design an algorithm called the cost-sensitive active learning through statistical methods (CATS). Experiments were performed on 12 binary-class datasets for both the distribution evaluation and learning task. The results of significance tests verify the effectiveness of CATS and its superior performance with respect to state-of-the-art cost-sensitive active learning algorithms. [ABSTRACT FROM AUTHOR]

• A novel privacy preserving method for data publication is proposed based on conditional probability distribution and machine learning techniques, which can offer stronger privacy guarantee and retain better data utility. • A basic algorithm and a complete algorithm are designed respectively for the settings of single sensitive attribute and and multiple sensitive attributes. • An improved complete algorithm is developed using Gibbs sampling, in order to advance the data utility when data are not sufficient. • Both theoretical analysis and experimental results are given, which validate the proposed method. Privacy has received increasing concerns in publication of datasets that contain sensitive information. Preventing privacy disclosure and providing useful information to legitimate users for data mining are conflicting goals. Generalization and randomized response methods were proposed in database community to tackle this problem. However, both of them have postulated the same prior belief for all transactions, which might be wrong modeling and lead to privacy breach. Besides, generalization and randomized response methods usually require a privacy controlling parameter to control the tradeoff between privacy and data quality, which may put the data publishers in a dilemma. In this paper, a novel privacy preserving method for data publication is proposed based on conditional probability distribution and machine learning techniques, which can achieve different prior beliefs for different transactions. A basic cross sampling algorithm and a complete cross sampling algorithm are designed respectively for the settings of single sensitive attribute and multiple sensitive attributes, and an improved complete algorithm is developed by using Gibbs sampling, in order to enhance data utility when data are not sufficient. Our method can offer stronger privacy guarantee, while, as shown in the extensive experiments, retaining better data utility. [ABSTRACT FROM AUTHOR]

Sparsity prior is a powerful tool for compressive sensing reconstruction (CSR) of hyperspectral image (HSI). However, conventional HSI-CSR strategies are not tuned to extracting refine spatial and spectral sparsity prior. Moreover, these CSR techniques are weak in preserving edges and suppressing artifacts. To alleviate these issues, this paper represents a first effort to characterize the spatial and spectral knowledge using the structure-based sparsity prior. Specifically, we introduce the nonlocal low-rank matrix recovery model and the hyper-Laplacian prior to encode the spatial and spectral structured sparsity, respectively. The key advantage of the proposed method, termed as hyper-Laplacian regularized nonlocal low-rank matrix recovery (HyNLRMR), is to adopt insightful property, namely the nonlocal self-similarity across the spatial domain and the consistency along the spectral domain. Then, the alternative direction multiplier method (ADMM) is designed to effectively implement the proposed algorithm. Experimental results on various HSI datasets verify that the proposed algorithm can significantly outperform existing state-of-the-art HSI-CSR methods. [ABSTRACT FROM AUTHOR]

• We propose G-SMOTE a new oversampling algorithm for imbalanced learning. • G-SMOTE can be seen as a geometric generalization of the SMOTE algorithm. • G-SMOTE defines a flexible geometric region around each minority class instance. • The tests performed show that G-SMOTE significantly outperforms SMOTE. • The G-SMOTE implementation is open source and it compatible to Scikit-Learn. Classification of imbalanced datasets is a challenging task for standard algorithms. Although many methods exist to address this problem in different ways, generating artificial data for the minority class is a more general approach compared to algorithmic modifications. SMOTE algorithm, as well as any other oversampling method based on the SMOTE mechanism, generates synthetic samples along line segments that join minority class instances. In this paper we propose Geometric SMOTE (G-SMOTE) as a enhancement of the SMOTE data generation mechanism. G-SMOTE generates synthetic samples in a geometric region of the input space, around each selected minority instance. While in the basic configuration this region is a hyper-sphere, G-SMOTE allows its deformation to a hyper-spheroid. The performance of G-SMOTE is compared against SMOTE as well as baseline methods. We present empirical results that show a significant improvement in the quality of the generated data when G-SMOTE is used as an oversampling algorithm. An implementation of G-SMOTE is made available in the Python programming language. [ABSTRACT FROM AUTHOR]

• This paper proposes a blockchain-based model that can protect users' privacy. • The ledger is variant that can adapt the computational power of users' terminals. • The model uses fewer resources compared with traditional blockchain systems. • The model owns better data utility than different privacy. • The model has a lower communication cost than some newest techniques. Social networks record a significant amount of user behavior data every day. By analyzing this behavior data, companies or attackers use it for marketing or more questionable purposes. We propose a blockchain-based model to protect the privacy of users' data in such big data environments. Traditional blockchain methods require too many resources for this task, so we propose a model that secures sensitive user information in a distributed blockchain and passes nonsensitive information through to the primary system in order to manage the blockchain size. Our DEPLEST algorithm performs these synchronization operations to keep local database storage and computational capacity within the limits of individual users' devices. We also propose a consensus protocol for blockchain ledger maintenance that runs well on typical client systems and prove that this protocol has excellent Byzantine fault tolerance (BFT). Our experimental results show that DEPLEST meets the architectural and performance needs and that our consensus protocol outperforms the existing proof of work (PoW) and proof of stake (PoS) methods in this application. [ABSTRACT FROM AUTHOR]

• Propose a Bayesian Possibilistic C-Means clustering for missing attribute estimation. • The proposed BPCM can discover the underlying patterns in data. • Propose a fuzzy ensemble learning scheme for cervical cancer screening. Recently, a lot of attention has been given to the treatment of cervical cancer due to its high lethality and morbidity. Early screening of this disease is of vital importance. In this paper, we propose an automatic cervical cancer screening algorithm that analyzes the related risk factors to provide preliminary diagnostic information for medical practitioners. In cervical cancer screening, a number of risk factors are considered to be highly private or sensitive, and many patients elect not to provide the corresponding information. Such severe amount of missing attributes leads to great difficulties for many automatic screening algorithms. To solve this problem, a Bayesian Possibilistic C-means (BPCM in short) clustering algorithm is proposed to discover the representative patterns from the complete data and to estimate the missing values of a specific sample using its closest representative pattern. After the data completion step, a two-stage fuzzy ensemble learning scheme is proposed to derive the final screening result. In the first stage, the bootstrap aggregation (bagging in short) procedure is adopted to sample the entire class-imbalanced dataset into a number of class-balanced subsets. In the second stage, a number of weak classifiers are trained on each subset and a fuzzy logic based approach is designed to analyze the classification results of the weak classifiers and to obtain the final classification result. Experiments have been conducted on a dataset containing 858 patients. From the experiment results, it can be observed that the proposed BPCM can effectively discover the underlying patterns and is reliable in estimating the missing attribute compared with the traditional approaches. Moreover, by applying the proposed fuzzy ensemble learning scheme, the final classification results on the completed data by BPCM are promising (an accuracy of 76% or a positive sensitivity of 79%) under the severe missing-attribute scenario (only 6% samples with complete data). [ABSTRACT FROM AUTHOR]

Hesitant fuzzy linguistic preference relations (HFLPRs) can well denote the decision makers (DMs)' hesitant qualitative comparison judgments. Considering that consistency and consensus are two key issues in group decisionmaking (GDM), this paper discusses GDM with HFLPRs within the context of multiplicative consistency and consensus. First, a multiplicative consistency concept for linguistic preference relations (LPRs) is presented, and a method for obtaining the priority weight vectors of LPRs is offered. Then, a method of deriving the priority weight vector of a HFLPR is proposed. Furthermore, the consistency index of a HFLPR is defined, an acceptable multiplicative consistency definition for HFLPRs is offered. Subsequently, for a HFLPR which has unacceptable consistency, an algorithm to improve its consistency level is developed. With respect to GDM with HFLPRs, a method to calculate the DMs' weights is provided. Then, individual HFLPRs are integrated into a collective HFLPR. After that, a consensus index is defined, and an algorithm to improve the consensus is proposed that can guarantee the multiplicative consistency of individual HFLPRs. Moreover, an automatic algorithm to GDM with HFLPRs is developed. Lastly, we apply the new methods to two practical problems to verify their operability and applicability. [ABSTRACT FROM AUTHOR]

Clustering streams of data is of increasing importance in many applications. In this paper, we propose a new synchronization-based clustering approach for evolving data streams, called SyncTree, which maintains all micro-clusters at different levels of granularity depending upon the data recency. Instead of using a sliding window or decay function to focus on recent data, SyncTree summarizes all continuously-arriving objects as synchronized micro-clusters sequentially in a batch fashion. Owing to the powerful concept of synchronization, the derived micro-clusters truly reflect the intrinsic cluster structure rather than summarize statistics of data, and old micro-clusters can be intuitively summarized at a higher level by iterative clustering to fit memory constraints. Building upon the hierarchical micro-clusters, SyncTree allows investigating the cluster structure of the data stream between any two time stamps in the past, and also provides a principled way to analyze the cluster evolution. Empirical results demonstrate that our method has good performance compared to state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Cloud computing is a new storage and calculation mode, which has been widely used as a tool to store and analyze users' data. With the development of industrial intelligence and big data, it is also prevalent in industrial field to accelerate production or to discover more knowledge to make better decisions. There are three main properties in cloud computing, namely dynamic and time-variation, large-scale, and change-of-ownership. Therefore, the data security problems are different and complicated than those in conventional networks. The existing methods, such as encrypted storage, security audit, and access control, cannot solve these security problems proactively and efficiently due to their own drawbacks. Commonly, self-samples always have similar features, while non-self-samples are abnormal in their own way. This paper mainly focuses on these two sources of data, one is intrusion detection of external non-self-samples and another is security calculation of self-samples. In this case, an improved dynamic immune algorithm (IDIA) is proposed, which is composed of an improved negative selection algorithm (iNSA) with the way of shift mutation and an improved dynamic clone selection algorithm (iDCS) with random grouping strategy. The former algorithm is used to generate detectors, and the latter is used to update them dynamically and adaptively. On this basis, an automated industrial production system is designed to guarantee the data security that not been accessed by external or other users. Datasets of KDD CUP99 and real industrial production data are collected to evaluate the performance of IDIA. The experiment results show the proposed algorithm can detect the intrusive data samples and identify autologous ones efficaciously. [ABSTRACT FROM AUTHOR]

With the development of cloud computing and the rise of smart city, smart city cloud service platforms are widely accepted by more and more enterprises and individuals. The underlying cloud workflow systems accumulate large numbers of business process models. How to achieve efficiently querying large process model repositories in smart city cloud workflow systems is challenging. To this end, this paper proposes an improved two-phase retrieval approach for querying large process model repositories in smart city cloud workflow systems. In the filtering stage, the index based on quantitative ordering relation with time and probability constraints (namely ORTP_index) is adopted to greatly reduce the number of candidate models in large process model repositories. In the refining phase, a process behavior similarity computing algorithm based on quantitative ordering relations is proposed to refine the candidate model set. Experiments illustrate that our proposal can significantly improve the query efficiency of large process model repositories in smart city cloud workflow systems based on behavior. [ABSTRACT FROM AUTHOR]

In this paper, we psropose a novel method for construction of a distance function and demonstrate its application in image segmentation. In algorithms for image segmentation, distance functions represent a criterion which divides pixels into groups of segments. We introduce two extended aggregation functions, extended powers product and extended weighted arithmetic mean of powers. Their relevant properties are examined, as well as certain resulting properties of distance functions, which are constructed by an application of mentioned aggregation functions. In addition, one pixel descriptor, which is motivated by Local Binary Pattern family of descriptors (LBP s), is introduced and discussed. In the experimental section, we present an application of the introduced extended aggregation functions and descriptor, by a construction of a new distance function, used in Fuzzy c-Means Clustering Algorithm (FCM) for image segmentation. [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]

• A novel general twin support vector machine with pinball loss is proposed. • The proposed algorithm is less sensitive to noise and more stable for re-sampling. • The proposed algorithm is efficient in terms of computation complexity. • Numerical results have shown better generalization performance for noise corrupted datasets. The standard twin support vector machine (TSVM) uses the hinge loss function which leads to noise sensitivity and instability. In this paper, we propose a novel general twin support vector machine with pinball loss (Pin-GTSVM) for solving classification problems. We show that the proposed Pin-GTSVM is noise insensitive and more stable for re-sampling. Further, the computational complexity of the proposed Pin-GTSVM is similar to that of the TSVM. Thus, the pinball loss function does not increase the computation time of the proposed Pin-GTSVM. Numerical experiments with different noise are performed on 17 UCI and KEEL benchmark real-world datasets and the results are compared with other baseline methods. The comparisons clearly show that the proposed Pin-GTSVM has better generalization performance for noise corrupted datasets. [ABSTRACT FROM AUTHOR]