Clustering Web documents using a combination of simulated annealing algorithm based on user behavior

Clustering Web documents using a combination of simulated annealing algorithm based on user behavior

saeed mostafapour gandali1

1) A graduate student in computer software trends, Islamic Azad University of Abadan, Iran -

IOI Code : XYHD-BEKGZ
Publication : International Conference On Research Science And Technology(restconf.org)
Abstract :
    At the present time the most important source of information the web pages found on the Internet That these pages are increasingly on the rise. On the other hand users to reach your target pages, They are forced to spend too much time searching the Internet. There has been much research on the classification and Clustering Web pages have done. Existing approaches to clustering are intrinsic to the documents. To We believe that users interact with information systems contain useful. For example, if two users with similar tastes Document access, represents the similarity of two documents. This can help content based on documents Other properties in addition to their intrinsic properties) characteristics of user behavior (clustered. In this project effort This field is and try to extract information using user behavior, more information and better with costs Less computationally extracted from web pages and documents to be used in clustering, so in this study, Clustering and simulated annealing algorithm combined efforts to reduce the computational cost and accuracy of clustering documents To increase. The main reason is optimized clustering method to select optimal sites for the cluster. The results of this work Research shows good efficiency in detection accuracy in clustering web pages as well as reduce costs Computational clustering Web pages show

Keywords : user behavior clustering simulated annealing algorithm restarting