Collaborative Recommendation via Adaptive Association Rule Mining

Collaborative recommender systems allow personalization for e-commerce by exploiting similarities and dissimilarities among users' preferences. We investigate the use of association rule mining as an underlying technology for collaborative recommender systems. Association rules have been used with success in other domains. However, most currently existing association rule mining algorithms were designed with market basket analysis in mind. Such algorithms are ine cient for collaborative recommendation because they mine many rules that are not relevant to a given user. Also, it is necessary to specify the minimum support of the mined rules in advance, often leading to either too many or too few rules; this negatively impacts the performance of the overall system. We describe a collaborative recommendation technique based on a new algorithm speci cally designed to mine association rules for this purpose. Our algorithm does not require the minimum support to be speci ed in advance. Rather, a target range is given for the number of rules, and the algorithm adjusts the minimum support for each user in order to obtain a ruleset whose size is in the desired range. Rules are mined for a speci c target user, reducing the time required for the mining process. We employ associations between users as well as associations between items in making recommendations. Experimental evaluation of a system based on our algorithm reveals performance that is signi cantly better than that of traditional correlationbased approaches. Corresponding author. Present a liation: Department of Computer Science, Wellesley College, 106 Central Street, Wellesley, MA 02481 USA, e-mail: salvarez@wellesley.edu

[1]  Y. Shoham,et al.  Ecom Syst Content-based, Collaborative Recommendation , 1997 .

[2]  Jaideep Srivastava,et al.  Web mining: information and pattern discovery on the World Wide Web , 1997, Proceedings Ninth IEEE International Conference on Tools with Artificial Intelligence.

[3]  Rajeev Motwani,et al.  Dynamic itemset counting and implication rules for market basket data , 1997, SIGMOD '97.

[4]  John Riedl,et al.  GroupLens: an open architecture for collaborative filtering of netnews , 1994, CSCW '94.

[5]  Geoffrey I. Webb.  Efficient search for association rules , 2000, KDD '00.

[6]  AgrawalRakesh,et al.  Mining quantitative association rules in large relational tables , 1996 .

[7]  Philip S. Yu,et al.  Efficient Data Mining for Path Traversal Patterns , 1998, IEEE Trans. Knowl. Data Eng..

[8]  Tomasz Imielinski,et al.  Mining association rules between sets of items in large databases , 1993, SIGMOD Conference.

[9]  Wynne Hsu,et al.  Integrating Classification and Association Rule Mining , 1998, KDD.

[10]  Sergio A. Alvarez,et al.  A New Adaptive-Support Algorithm for Association Rule Mining , 2000 .

[11]  Ramakrishnan Srikant,et al.  Mining quantitative association rules in large relational tables , 1996, SIGMOD '96.

[12]  Petr Hájek,et al.  On Generation of Inductive Hypotheses , 1977, Int. J. Man Mach. Stud..

[13]  Kristian J. Hammond,et al.  Mining navigation history for recommendation , 2000, IUI '00.

[14]  Ramakrishnan Srikant,et al.  Fast algorithms for mining association rules , 1998, VLDB 1998.

[15]  Pattie Maes,et al.  Social information filtering: algorithms for automating “word of mouth” , 1995, CHI '95.

[16]  Michael J. Pazzani,et al.  Learning Collaborative Information Filters , 1998, ICML.