Search

Abstract: The partitional clustering algorithms are used more widely in text clustering. However, the traditional algorithms based on partition treat all the attributes equally in clustering process. They all suppose that the importance of each attribute is equal. These algorithms will have a lower accuracy. In order to handle this problem, this paper provides a new clustering algorithm-attribute weighted fuzzy c-means algorithm. During the iteration of this algorithm, it can find the important attributes. Moreover, this algorithm also can find the cluster structure hiding by the unimportant attributes. The simulation of this algorithm on test documents can prove the algorithm provided by this paper can gain a good computation speed, found the latent structure and can remark the different importance of each attribute. [Copyright &y& Elsevier]

Abstract: With the rapid development of the Intemet, information overload and isotropic becomes worse and worse. Personalized services system''s birth partly resolved this problem. The traditional recommendation methods, such as content based recommendation and collaborative filtering, do help a lot. However, to some extent, it couldn’t authentically understand the preference of users. Considered the limitations of the traditional methods, tag recommedation spring up. This paper makes expanded research based on tag reccommendation. With semantic matching, a certain tag turns to a similar tag set. Acording to co-occurence of tags, a tag is expended to a ralated tag set. Integrate these tag set to generate recommendation list is better than single tag recomendation by experimental observation. Besides, this paper also adopt time degradatioin aglorithm which improved the recommender''s accuracy and efficiency. [Copyright &y& Elsevier]

Abstract: In this paper, we address the problem of consistency checking for Point Objects. A coordinate graph representation is introduced. The basic idea is to project the spatial constraints on both X and Y dimensions, and the graph is constructed on each dimension. By using the graph representation, the problem of consistency checking is then transformed to a graph cycle detection problem. The consistency checking can be achieved with O(N+E) time as well as space complexity, where N is the number of spatial objects, and E is number of spatial predicates in the constraint. The proposed approach to consistency checking for point objects is faster than O(N2) when the number of predicates is much smaller than N2 . The algorithm can also guarantee the global consistency. [Copyright &y& Elsevier]

Abstract: Since the estimation of distribution algorithm(EDA) was introduced, different approaches for continuous problems solving have been developed. Through the efforts of many scholars, the EDAs brought some success in continuous domain, but there are still some deficiencies: Gaussian-based EDA doesn’t perform well in problems with many optima, such as Schwefel function; histogram-based EDA doesn’t perform well in problems with singular global optimum, such as Three-Peak function. Because of this, based on improved histogram-based EDA and referred to the idea of k-Means clustering, we built cooperative EDA. Experimental results showed that the improved algorithm in this paper can give comparable with or better performance than those improved algorithms. [Copyright &y& Elsevier]

Abstract: In this paper, a kind of nonlinear optimization problems with nonlinear equality and inequality constraints are discussed, and a new algorithm with arbitrary initial point is presented by combining the norm-relaxed method of feasible direction for inequality constrained optimization with the idea of strongly sub-feasible direction method and the two concepts of quasi-KKT point and quasi-multiplier method. Firstly, the original problem is transformed into an associated simpler problem with only inequality constraints and a parameter. The later problem is shown to be approximately equivalent to the original problem if the parameter is large enough (but finite). At each iteration of the proposed algorithm, a direction is obtained by solving one direction finding subproblem which always possesses a solution. Some good properties of the new algorithm are that the initial iteration point is arbitrary and the number of the constraint functions satisfying the inequality constraints is nondecreasing. The algorithm is proved to be globally and strongly convergent under mild conditions. Finally, the numerical experiments are implemented on Matlab7.0. [Copyright &y& Elsevier]

Abstract: To resolve the problems such as how to implement 3D CAD model retrieval and how to reuse the retrieval results, this paper descripes an ontology-based algorithm for mechanical part retrieval. The mechanical part is segmented into many relevant sub-parts. And then, attach semantic descriptions and annotations to these sub-parts. Finally, evaluate the similarity of the models based on the semantic ontology. This method can efficiently achieve the similarity assessment of the mechanical parts and as well reflect human perceptual similarity. [Copyright &y& Elsevier]

Abstract: Clustering is an example of a class of optimization problems. In the classical clustering, an item must belong to any one cluster. But fuzzy clustering describes more accurately the ambiguous type of structure in data. The fuzzy ISODATA clustering exhibits the rapid convergence in finding the best classification program when the classification number is given. In this paper, we propose the algorithm to solve the choosing lands problem and show the result of the experiment. The result is satisfied to realtors in choosing lands. [Copyright &y& Elsevier]

Abstract: In this paper, the Multi-Depot Vehicle Routing Problem (MDVRP), a generalization of the Vehicle Routing Problem (VRP) is reviewed. We mainly review the MDVRP from the following several aspects: the definition, solution methods (including exact algorithms, heuristics and meta-heuristics) and problem variants. Our objective is to survey the MDVRP systematically and draw more attention to it from researchers and practitioners. [Copyright &y& Elsevier]

Abstract: Large numbers of visualization services exist in Web, bringing about service discovery problems for those who performs visualization tasks. An approach to visualization services selection based on visualization characteristics is proposed in this paper. Firstly, an architecture to discover services in visualization domain is proposed. Secondly, a novel algorithm of visualization services selection based on visualization characteristics is discussed. Finally, a case study is given to review the availability of the algorithm. [Copyright &y& Elsevier]

Abstract: Clustering is a method of unsupervised learning, and a common technique for intelligent data analysis used in many fields. The enlarging volumes of information produced by the progress of technology, makes clustering of very large scale data a challenging task. Therefore, parallel processing of very large scale data set is a good solution to that problem. CLARANS is approved to be an efficient and effective clustering method for spatial data mining. It is insensitive to outliers and has the advantage of processing category data directly compared to other partition method like K-means. However, it is a serial algorithm and cannot process large scale data set efficiently. In this paper, we propose a parallel CLARANS clustering algorithm based on MapReduce, which is a current and powerful parallel programming framework. Experimental results demonstrate that the parallel CLARANS algorithm can handle large scale data set efficiently and effectively. Furthermore, the evaluation criterions of scaleup, speedup and sizeup validate its efficiency. [Copyright &y& Elsevier]

Abstract: In ISAR imaging of the target with rotating parts, the motion compensation algorithm is usually invalid because the disturbance of rotating parts. To solve this problem, this paper proposes a novel algorithm based on dividing the data into several pieces. The range profile is divided into several parts, then the interesting part is selected by the minimum entropy criterion, finally the influence of rotating parts is restrained effectively, and well-focused imaging can be obtained. This is very important for the separating rotating parts from the rigid body. Imaging results of both simulated and real data are given to verify the validity of the proposed algorithm. [Copyright &y& Elsevier]

Abstract: The detection of small targets that submerges in the homogeneous image background is a difficult problem in the design of small targets detection algorithm. The concept of local reverse entropy combining the concepts of local ntropy with reverse entropy is proposed in this paper, which is to solve that problem by enhancing small targets. The norms of local signal-to-background ratio and elapsed time are used to demonstrate the performance of local reverse ntropy map. Both quantitative analysis and qualitative comparison confirm the validity and efficiency of the resented approach. [Copyright &y& Elsevier]

Abstract: Fuzzy association rules are easily understood by users. Simulated annealing algorithm is an effective approximation algorithm which is suitable for solving large-scale combinatorial optimization problem. In this paper we proposed a classifier design method based on simulated annealing algorithm and fuzzy association rules. It can achieve satisfied accuracy and a small number of classification rules. Experiments demonstrated the effectiveness and advantages of the proposed algorithm. [Copyright &y& Elsevier]

Abstract: Optimized configuration of energy industry is an issue for resource combination and optimization, which includes continuous variable and discrete variable and belongs to the optimal power flow research. In order to improve the solution quality and the counting efficiency, this paper gives the solution using the mixed Greedy Algorithm and Genetic Algorithm: Use the superiority of Greedy Algorithm to obtain the original resource distribution solution space in the structure. This algorithm can minimize the spatial scale. And then utilizes the Genetic Algorithm to select the original solution space based on the overall situation and improve its application efficiency. The test case matches the anticipation. [Copyright &y& Elsevier]