On Enumerating Maximal Bicliques of Click-Through Graph

Main Article Content

G. Manoranjitham
A. Revathi

Abstract

Search Engine is to retrieve the relevant information from the web, based on user queries. The problem of discovering query clusters from a click-through graph of web search logs is described in this paper. The click through graph consists of a set of clicked queries, pages that are returned as result and a set of edges that connects a query and a page clicked by the user. The proposed method is enumerating all maximal bicliques of a bipartite graph from click-through graph and compute equivalence query clusters from maximal bicliques. It has two main contributions: first, use all maximal bicliques to find minimum number of bicliques covering a subset of edges or vertices for given bipartite graph. We have conducted experiments on Yahoo web search queries.


Keywords: biclique enumeration; click-through graph; query cluster;

Downloads

Download data is not yet available.

Article Details

Section
Articles