A New Method of Clustering Search Results Using Frequent Itemsets with Graph Structures

摘要

The representation of search results from the World Wide Web has received considerable attention in the database research community. Systems have been proposed for clustering search results into meaningful semantic categories for presentation to the end user. This paper presents a novel clustering algorithm, which is based on the concept of frequent itemsets mining over a graph structure, to efficiently generate search result clusters. The performance study reveals that the algorithm was highly efficient and significantly outperformed previous approaches in clustering search results.