Search

Abstract: To deal with the file pollution problem in structured P2P networks, this paper proposes an anti-pollution strategy based on the immune approach and the file propagation-chain tracing technique. With the characteristics of the structured P2P networks, we establish a biological immune model and a file propagation tracing mechanism, based on which the mechanisms for alarms collection, propagation-chain tracing, immune response and antibody memorizing are designed. The proposed immune strategy can not only eliminate the polluted clusters and indices effectively but also prevent the reviving of the old pollution. In addition, the node permission is employed to control the ability to request files. The experimental results show that our strategy can curb the propagationsof the polluted files effectively and efficiently even in the severely polluted structured P2P networks. [Copyright &y& Elsevier]

Abstract: In previous works, the agent communication always takes place randomly through the whole agents distributed area, and all agents are treated equally in the communication process. However, the real communication is always imbalanced: communication in some places of the system may be very busy and the communication resources are insufficient, whereas in other places the communication may be very idle and the communication resources are redundant. To make up the shortage of related works, this paper presents a cluster partition-based communication model for multiagents. In the presented model, agents are divided into some clusters according to their communication frequencies; in each cluster, one agent is set to be hub-agent which carries the information of other agents within the cluster. With the presented model, the communication costs within the same cluster is lower than the one among different clusters because agents have closer distances within the same cluster; moreover, many real communication processes may take place within clusters since the clusters are partitioned according to the communication frequencies among agents. Therefore, the communication time and costs can be reduced by adopting our model. Finally, the advantage of our model is proved by a series of experiments. [Copyright &y& Elsevier]

Abstract: Radio-frequency identification (RFID) uses non-contact data access mechanism to read or write tag data through the radio frequency transmitted by the readers. Currently, the RFID systems are applied in different areas. However, the data access rate of a RFID system could be interfered by the environment furnishings, barriers, deployment method, and number of readers. The system stability heavily depends on the experiences of the deployment engineers especially for the active readers. This paper proposes an intelligent deployment system for RFID readers. The system uses the furnished simulation of environment and the best deployment of readers to construct an automatic deployment system for the RFID readers. The former constructs a fuzzy interference model of barriers according to the materials and inference degree of the furnishings. The latter uses the reader cluster and genetic algorithm to establish the best deployment plan of readers. The experimental results show that the system can reduce the interference problem of reader deployment as well as improve the reliability and stability of the RFID systems. [Copyright &y& Elsevier]

Abstract: Clustering problem is an unsupervised learning problem. It is a procedure that partition data objects into matching clusters. The data objects in the same cluster are quite similar to each other and dissimilar in the other clusters. Density-based clustering algorithms find clusters based on density of data points in a region. DBSCAN algorithm is one of the density-based clustering algorithms. It can discover clusters with arbitrary shapes and only requires two input parameters. DBSCAN has been proved to be very effective for analyzing large and complex spatial databases. However, DBSCAN needs large volume of memory support and often has difficulties with high-dimensional data and clusters of very different densities. So, partitioning-based DBSCAN algorithm (PDBSCAN) was proposed to solve these problems. But PDBSCAN will get poor result when the density of data is non-uniform. Meanwhile, to some extent, DBSCAN and PDBSCAN are both sensitive to the initial parameters. In this paper, we propose a new hybrid algorithm based on PDBSCAN. We use modified ant clustering algorithm (ACA) and design a new partitioning algorithm based on ‘point density’ (PD) in data preprocessing phase. We name the new hybrid algorithm PACA-DBSCAN. The performance of PACA-DBSCAN is compared with DBSCAN and PDBSCAN on five data sets. Experimental results indicate the superiority of PACA-DBSCAN algorithm. [Copyright &y& Elsevier]

Abstract: Large amounts of high-dimensional data are distributed with the application of networks. Distributed clustering has become an increasingly important task due to variety of real-life constrains, including bandwidth and security aspects. Many distributed clustering algorithm have been proposed, but most of them have high transmission cost and poor clustering quality. In this paper, we propose a scalable local density-based distributed clustering algorithm which can easily fit high-dimensional data sets by this method such as density attractor distance and noise factor. In order to keep a lower transmission cost, we determine suitably low factor noises to send to the server. Furthermore, Test data sets, CMC data sets and KDD-CUP-99 are used for experimental evaluation to validate the performance practically. The experimental results and theoretical analysis show that the efficiency and quality for clustering of the proposed algorithm are superior to the other distributed clustering algorithm. [Copyright &y& Elsevier]

Abstract: The support vector clustering (SVC) algorithm consists of two main phases: SVC training and cluster assignment. The former requires calculating Lagrange multipliers and the latter requires calculating adjacency matrix, which may cause a high computational burden for cluster analysis. To overcome these difficulties, in this paper, we present an improved SVC algorithm. In SVC training phase, an entropy-based algorithm for the problem of calculating Lagrange multipliers is proposed by means of Lagrangian duality and the Jaynes’ maximum entropy principle, which evidently reduces the time of calculating Lagrange multipliers. In cluster assignment phase, the kernel matrix is used to preliminarily classify the data points before calculating adjacency matrix, which effectively reduces the computing scale of adjacency matrix. As a result, a lot of computational savings can be achieved in the improved algorithm by exploiting the special structure in SVC problem. Validity and performance of the proposed algorithm are demonstrated by numerical experiments. [Copyright &y& Elsevier]

Abstract: This paper presents selective regeneration particle swarm optimization (SRPSO), a novel algorithm developed based on particle swarm optimization (PSO). It contains two new features, unbalanced parameter setting and particle regeneration operation. The unbalanced parameter setting enables fast convergence of the algorithm and the particle regeneration operation allows the search to escape from local optima and explore for better solutions. This algorithm is applied to data clustering problems for performance evaluation and a hybrid algorithm (KSRPSO) of K-means clustering method and SRPSO is developed. In the conducted numerical experiments, SRPSO and KSRPSO are compared to the original PSO algorithm, K-means, as well as, other methods proposed by other studies. The results demonstrate that SRPSO and KSRPSO are efficient, accurate, and robust methods for data clustering problems. [Copyright &y& Elsevier]