Semantic Log Analysis Based on a User Query Behavior Model

  • Authors:

  • Kawamae Noriaki;Mukaigaito Takeya;Hanaki Miyoshi

  • Affiliations:

  • -;-;-

  • Venue:
  • ICDM '03 Proceedings of the Third IEEE International Conference on Data Mining
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose a novel log analysis method to capture thesemantic relations among words appearing in Web searchlogs. Our method focuses on the reciprocal relations amonga user's intentions, stages of information need, and querybehavior in seeking information via a search engine. Theapproach works because it is based on the assumption that auser's intentions in each query can be derived as a model onthe basis of his stage of information need and query behavior,through multiple empirical observations of search logs.The user's intentions drive user to change the words in eachsuccessive queries and can thus be used to clarify the semanticrelations among words. As a result, this method hasthe advantage of capturing the semantic relations amongwords without requiring either manual or natural languageprocessing. Our experimental results indicate that semanticrelations could successfully be derived from search logs,confirming that an ontology and thesaurus could be constructedautomatically.