关键词: 基于密度的聚类算法, 网页相似度, 聚类, 冗余网页

Abstract: Nowadays there are many redundancy pages in results of search engine, and the results are not classified. An optimization algorithm of webpage search results based on an improved DBSCAN (density-based spatial clustering of applications with noise) algorithm is proposed and effective to cluster and classify the results. The algorithm selects the webpages with search weights above a certain value from all search results, then it extracts the eigenvalue of pages and candidate keys, compares the pages similarity to maximize the elimination of duplication and redundancy pages. Meanwhile, classifications are provided in accordance with the candidate keys of pages, thereby the precision and satisfaction of search engine could be improved with the effect of more intelligence.

Key words: DBSCAN algorithm, page similarity, clustering, redundancy page