Quantifying Asymmetric Semantic Relations from Query Logs by Resource Allocation

  • Authors:

  • Zhiyuan Liu;Yabin Zheng;Maosong Sun

  • Affiliations:

  • Department of Computer Science and Technology, State Key Lab on Intelligent Technology and Systems, National Lab for Information Science and Technology, Tsinghua University, Beijing, China 100084;Department of Computer Science and Technology, State Key Lab on Intelligent Technology and Systems, National Lab for Information Science and Technology, Tsinghua University, Beijing, China 100084;Department of Computer Science and Technology, State Key Lab on Intelligent Technology and Systems, National Lab for Information Science and Technology, Tsinghua University, Beijing, China 100084

  • Venue:
  • PAKDD '09 Proceedings of the 13th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper we present a bipartite-network-based resource allocation(BNRA) method to extract and quantify semantic relations from large scale query logs of search engine. Firstly, we construct a query-URL bipartite network from query logs of search engine. By BNRA, we extract asymmetric semantic relations between queries from the bipartite network. Asymmetric relation indicates that two related queries could be assigned different semantic relevance strength against each other, which is more conforming to reality. We verify the validity of the method with query logs from Chinese search engine Sogou. It demonstrates BNRA could effectively quantify semantic relations from We further construct query semantic networks, and introduce several measures to analyze the networks. BNRA is not only `language oblivious' and `content oblivious', but could also be easily implemented in a paralleled manner, which provides commercial search engines a feasible solution to handle large scale query logs.