Earticle

Time is an important dimension of information space. It plays important roles in Web search, because most Web pages contain time information and many Web queries are time-related. Therefore, exploiting temporal information in Web pages has been a hotspot in the research on Web search. In this paper, we focus on the time-enhanced topic clustering issue for news search results. Traditional clustering algorithms are usually based on the common phrases of Web pages, and they have little consideration about using the temporal information of Web pages. From this perspective, we propose a time-enhanced topic clustering algorithm for news Web pages. It improves traditional algorithms which only consider textual clustering, and applies a temporal clustering procedure on the topics returned by a textual clustering algorithm, which is to arrange every Web page in a cluster along a timeline based on the update time in Web pages. We conduct experiments on a real dataset crawled from Google News, and compare our algorithm with other competitors including K-Means, STC, TFIC, and Minhash Clustering in terms of different metrics such as precision and recall. The experimental results show that the proposed algorithm has better performance under both offline and online clustering test.

In power systems operation, state estimation takes an important role in security control. For the state estimation problem, the weighted least squares (WLS) method and the fast decoupled method have been widely used at present. Especially when bad data are mutually interacting, the detecting of multiple bad data may be difficult to handle, since the normalized or weighted residuals may become faulty. Then the problem of detecting bad data is considered as a combinatorial decision procedure. In this paper, the binary Particle Swarm Optimization (PSO) is used for the detecting of multiple bad data in the power system state estimation. The PSO, like other meta-heuristic algorithms, can handle constrains that would be troublesome in classical mathematical approach. However, population based algorithms require higher computing time to find optimal point. This shortcoming is overcome by a parallel processing of PSO algorithm. The parallel PSO algorithm is implemented on a PC cluster system with 8 personal computers. The proposed approach has been tested on the IEEE-14 and 118 bus systems. The results showed that the binary PSO based procedures behave satisfactorily in the detecting multiple bad data and computing time of parallelized PSO algorithm can be reduced without losing the quality of solution.

For further solving the problem of normative emergency plan, text mining should be combined with emergency plan compilation. Fixed " threshold strategy is used by traditional DBSCAN, which would lead to the problem of cluster boundary wrong recognition. Improved DBSCAN algorithm is introduced in this paper. Least Square Fit is taken to fit plans similarity curve to find the best of initial " threshold. According to the initial " , a new strategy is used to get dynamic " threshold to improve the precision and recall. The simulations results show that the presented method is efficient for providing intelligent reference groups for government staff.

Periodic pattern mining has gained a great attention in the past decade. Previous studies mostly focus on synchronous periodic patterns. The literature proposes many methods for mining periodic patterns. Nevertheless, asynchronous periodic pattern mining has gradually received more attention recently. In this paper, we propose an efficient 2-D linked structure and the OEOP (One Event One Pattern) algorithm to discover all kinds of valid segments in each single event sequence. Then, referring to the general model of asynchronous periodic pattern mining proposed by Huang and Chang, this study combines these valid segments found by OEOP into 1-patterns with multiple events, multiple patterns with multiple events and asynchronous periodic patterns. The experimental results show that these algorithms have good performance and scalability.

Spatial Partitioning Fragmentation (SPF) is a popular method to partition data in Distributed Spatial Databases (DSDBs). The issue of cross-border queries is an inherent problem however with distributed spatial data queries based on partitioning fragmentation given a continuity and strong correlation of geospatial data. In the case of partitioning fragmentation, a global spatial join can be translated into multiple sub-joins, and then divided into 2 groups: Cross-Border Joins (CBJs) and Non-Cross-Border Joins (NCBJs). The CBJ approach is essential for process efficiency in a distributed spatial query. A compound join based on a topological relationship inquiry and a buffering analysis is a crucial class of spatial queries. This article studies compound join optimization for spatial queries in a DSDB, and proposes a set of theorems and rules for the optimization of CBJs, contributing a removal rule and a filtering rule. This article supplies a Partition Fragmentation Join Strategy (PFJS) to resolve the compound join problem based on these rules. Experimental results show that the PFJS can improve the efficiency of CBJs, when compared with the Naive Join Strategy (NJS) or the Spatial Semi-Join Strategy (SSJS). The PFJS contributes to the optimization of spatial compound joins.

The need for providing mobile multimedia services is increasing as the mobility of individuals’ increases and wireless network are widely adopted. The advantages of Cloud Computing is utilized to address the shortage in computation, memory, and energy resources to support advanced graphic processing functions in real-time, especially for high-resolution digital multimedia contents, as well as security vulnerabilities for wireless networks connecting mobile terminals. This paper presents a framework for streaming digital multimedia contents through cloud computing infrastructure. The framework describes the relationship among the service provider, the cloud provider, and the mobile device. The main aim of this framework is to provide a scheme for multimedia streaming service that can operate in a ubiquitous environment through mobile devices.

To considered the problem of discovering of interesting association rule among item sets in data base. Algorithms for mining association rule are practical methods to find interesting rules implied in large database. Proposed an innovative approach, beyond minimum support and minimum confidence framework, some extra measures consider for rule improvement is user interestingness constraint. It use three user defined constraint minimum support, minimum confidence, and interesting item and in addition makes use of k- nonempty subset generation methodology of the item which are user interest. Proposed algorithm fundamentally different from the identified algorithms, a number of algorithm is developed for association rule mining, the identified algorithms go through as of number of scanning of data base, and generate the candidate item set , unnecessary or uninteresting rule . The current method applies the user interesting constraint to generate only interesting association rule in data base. Proposed approach not just reduces the number of scanning of data base but also generated frequent itemset, and mine interesting association rule. Experimental result shows that the number of uninteresting rules can be reduced successfully and the validity of rules which mined are better.

Information retrieval systems and Web search engines apply highly optimized techniques for compressing inverted indices. These techniques reduce index sizes and improve the performance of query processing that uses compressed indices to find relevant documents for the users' queries. In this paper, we proposed a novel technique for querying compressed Arabic inverted indices in search engines. The technique depends on encoding Arabic terms stored in the inverted indices of Web search engines and information retrieval systems. This minimizes the storage space required for those terms in the index and decrease the number of comparisons needed for query matching in the query processing stage. The number of comparisons is decreased by minimizing the number of bytes that represent the terms in the index and applying the same encoding technique to the keywords of the query. The results showed a 38% reduction in total size of the index. The average number of comparisons to find a word is also decreased in the new index. Both sequential and binary searches were decreased by: 13.58%, and 38.63% respectively relative to the total number of comparisons of each keyword in the query.

Recently Burusco introduced interval-valued fuzzy formal contexts into fuzzy formal con- cept analysis. The most interesting work mainly including fuzzy attribute implications from fuzzy formal context, however, were presented under the framework of residuated lattice. In this paper, we rst show that the study of interval-valued fuzzy set can be tted into the framework of residuated lattice. Secondly, considering that the denition of fuzzy attribute implication in fact implies a minimal degree and thus may be impractical in some appli- cations, we introduce probability information to fuzzy inclusion degree and then to fuzzy attribute implication, and discuss some properties of this denition. The result veries the correctness of probability fuzzy attribute implication in some illustrations.