Using log mining to analyze user behavior on search engine

Abstract

Users’ behavior analysis has become one of the most important research topics, especially in terms of performance optimization, architecture analysis, and system maintenance, due to the rapid growth of search engine users. By adequately performing analysis on log data, researchers and Internet companies can get guidance to better search engines. In this paper, we perform our analysis based on approximately 750 million entries of search requests obtained from log of a real commercial search engine. Several aspects of users’ behavior are studied, including query length, ratio of query refining, recommendation access, and so on. Different information needs may lead to different behaviors, and we address this discussion in this paper. We firmly believe that these analyses would be helpful with respect of improving both effectiveness and efficiency of search engines.

Keywords

user behavior analysis / search engine / web information retrieval

References

[1] China Internet Network Information Center (CNNIC). The 25th report in development of Internet in China. 2010. http://www.cnnic.net.cn/uploadfiles/pdf/2010/1/15/101600.pdf

[2] Cockburn A, Jones S. Which way now? Analyzing and easing inadequacies in WWW navigation. International Journal of Human-Computer Studies, 1996, 45(1): 105–129

[3] Tauscher L, Greenberg S. How people revisit web pages: Empirical findings and implications for the design of history systems. International Journal of Human-Computer Studies, 1997, 47(1): 97–137

[4] Silverstein C, Marais H, Henzinger M, Moricz M. Analysis of a very large web search engine query log. ACM SIGIR Forum, 1999, 33(1): 6–12.

[5] Yu H, Liu Y, Zhang M, Ru L, Ma S. Research in search engine user behavior based on log analysis. Journal of Chinese Information Processing, 2007, 21(1): 109–114 (in Chinese)

[6] Agichtein E, Brill E, Dumais S. Improving web search ranking by incorporating user behavior information. In: Proceedings of the 29th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. 2006, 19–26

[7] Dou Z, Song R, Yuan X, Wen J. Are click-through data adequate for learning web search rankings? In: Proceedings of the 17th ACM Conference on Information and Knowledge Management.,2008, 73–82

[8] Liu Y, Cen R, Zhang M, Ru L, Ma S. Automatic search engine evaluation based on user behavior analysis. Journal of Software, 2008, 19(11): 3023–3032 (in Chinese)

[9] Sullivan D. Search engine sizes. 2005. http://searchenginewatch.com/reports/article.php/2156481

[10] Joachims T, Granka L, Pan B, Hembrooke H, Gay G. Accurately interpreting clickthrough data as implicit feedback. In: Proceedings of the 28th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. 2005, 154–161

[11] Downey D, Dumais S, Liebling D, Horvitz E. Understanding the relationship between searchers’ queries and information goals. In: Proceedings of the 17th ACM Conference on Information and Knowledge Mining. 2008, 449–458

RIGHTS & PERMISSIONS

2014 Higher Education Press and Springer-Verlag Berlin Heidelberg