Showing papers on "Dunn index published in 2010"
01 Jan 2010
TL;DR: This paper presents an ant-inspired method for clustering semantic Web services that considers the degree of semantic similarity between services as the main clustering criterion and proposes a matching method and a set of metrics to measure the semantic similarities between two services.
Abstract: This paper presents an ant-inspired method for clustering semantic Web services. The method considers the degree of semantic similarity between services as the main clustering criterion. To measure the semantic similarity between two services we propose a matching method and a set of metrics. The proposed metrics evaluate the degree of match between the ontology concepts describing two services. We have tested the ant-inspired clustering method on the SAWSDL-TC benchmark and we have evaluated its performance using the Dunn Index, the Intra-Cluster Variance metric and an original metric we introduce in this paper.
23 citations
05 Jul 2010
TL;DR: A new algorithm has been represented to cluster web pages based on data content based on the expressions and key words existed in web pages and their bit display a vector and using a new similarity criterion obtained from Cosine and Jaccard similarity criterion.
Abstract: Nowadays, using web and Internet as a world wide information system faces us with so many data. In this direction, the necessity of accessing some tools for data processing in web level which helps the man intelligently to transform these data into useful knowledge seems so important. Clustering the web pages is one of these techniques. In this paper, a new algorithm has been represented to cluster web pages based on data content. The new algorithm has been suggested based on the expressions and key words existed in web pages, and their bit display a vector and using a new similarity criterion obtained from Cosine and Jaccard similarity criterion. To evaluate the efficacy of suggested algorithm, some pages with five subjects of software engineering, computerized networks, architecture of computer, parallel processing and operating system have been investigated and after preparing a suitable data bed the represented algorithm has been simulated separately through two similarity criteria of Cosine and that of represented in this pager and has been evaluated using Dunn index. The results obtained from simulation show high efficiency of the algorithm proposed in separating web pages and their clustering. The represented algorithm can be used in most of the problems related to clustering web pages.
9 citations