Mining User preference using Spy voting for search engine personalization

This article addresses search engine personalization. We present a new approach to mining a user's preferences on the search results from clickthrough data and using the discovered preferences to adapt the search engine's ranking function for improving search quality. We develop a new preference mining technique called SpyNB, which is based on the practical assumption that the search results clicked on by the user reflect the user's preferences but does not draw any conclusions about the results that the user did not click on. As such, SpyNB is still valid even if the user does not follow any order in reading the search results or does not click on all relevant results. Our extensive offline experiments demonstrate that SpyNB discovers many more accurate preferences than existing algorithms do. The interactive online experiments further confirm that SpyNB and our personalization approach are effective in practice. We also show that the efficiency of SpyNB is comparable to existing simple preference mining algorithms.

[1]  Philip S. Yu,et al.  Partially Supervised Classification of Text Documents , 2002, ICML.

[2]  Wilfred Ng,et al.  Applying Co-training to Clickthrough Data for Search Engine Adaptation , 2004, DASFAA.

[3]  Avrim Blum,et al.  The Bottleneck , 2021, Monopsony Capitalism.

[4]  Garrison W. Cottrell,et al.  Automatic combination of multiple ranked retrieval systems , 1994, SIGIR '94.

[5]  Rakesh Agrawal,et al.  A framework for expressing and combining preferences , 2000, SIGMOD '00.

[6]  Clement T. Yu,et al.  Personalized Web search for improving retrieval effectiveness , 2004, IEEE Transactions on Knowledge and Data Engineering.

[7]  Jennifer Widom,et al.  Scaling personalized web search , 2003, WWW '03.

[8]  Taher H. Haveliwala.  Topic-sensitive PageRank , 2002, IEEE Trans. Knowl. Data Eng..

[9]  Thorsten Joachims,et al.  Making large scale SVM learning practical , 1998 .

[10]  Thomas G. Dietterich.  What is machine learning? , 2020, Archives of Disease in Childhood.

[11]  Clement T. Yu,et al.  Personalized web search by mapping user queries to categories , 2002, CIKM '02.

[12]  Masatoshi Yoshikawa,et al.  Adaptive web search based on user profile constructed without any effort from users , 2004, WWW '04.

[13]  Werner Kießling,et al.  Foundations of Preferences in Database Systems , 2002, VLDB.

[14]  Thorsten Joachims,et al.  Evaluating Retrieval Performance Using Clickthrough Data , 2003, Text Mining.

[15]  Hans Ulrich Simon,et al.  Robust Trainability of Single Neurons , 1995, J. Comput. Syst. Sci..

[16]  Wilfred Ng,et al.  Web dynamics and their ramifications for the development of Web search engines , 2006, Comput. Networks.

[17]  Jeffrey Heer,et al.  Separating the swarm: categorization methods for user sessions on the web , 2002, CHI.

[18]  Andrew McCallum,et al.  A comparison of event models for naive bayes text classification , 1998, AAAI 1998.

[19]  Alexander Pretschner,et al.  Ontology based personalized search , 1999, Proceedings 11th International Conference on Tools with Artificial Intelligence.

[20]  Thorsten Joachims,et al.  Accurately Interpreting Clickthrough Data as Implicit Feedback , 2017 .

[21]  Thorsten Joachims,et al.  Optimizing search engines using clickthrough data , 2002, KDD.

[22]  Philip S. Yu,et al.  Building text classifiers using positive and unlabeled examples , 2003, Third IEEE International Conference on Data Mining.