Applying Associative Retrieval Techniques to Alleviate the Sparsity Problem in Collaborative Filtering

Persistent Link:
http://hdl.handle.net/10150/105493
Title:
Applying Associative Retrieval Techniques to Alleviate the Sparsity Problem in Collaborative Filtering
Author:
Huang, Zan; Chen, Hsinchun; Zeng, Daniel
Citation:
Applying Associative Retrieval Techniques to Alleviate the Sparsity Problem in Collaborative Filtering 2004-01, 22(1):116-142 ACM Transactions on Information Systems
Publisher:
ACM
Journal:
ACM Transactions on Information Systems
Issue Date:
Jan-2004
Description:
Artificial Intelligence Lab, Department of MIS, University of Arizona
URI:
http://hdl.handle.net/10150/105493
Submitted date:
2004-08-16
Abstract:
Recommender systems are being widely applied in many application settings to suggest products, services, and information items to potential consumers. Collaborative filtering, the most successful recommendation approach, makes recommendations based on past transactions and feedback from consumers sharing similar interests. A major problem limiting the usefulness of collaborative filtering is the sparsity problem, which refers to a situation in which transactional or feedback data is sparse and insufficient to identify similarities in consumer interests. In this article, we propose to deal with this sparsity problem by applying an associative retrieval framework and related spreading activation algorithms to explore transitive associations among consumers through their past transactions and feedback. Such transitive associations are a valuable source of information to help infer consumer interests and can be explored to deal with the sparsity problem. To evaluate the effectiveness of our approach, we have conducted an experimental study using a data set from an online bookstore. We experimented with three spreading activation algorithms including a constrained Leaky Capacitor algorithm, a branch-and-bound serial symbolic search algorithm, and a Hopfield net parallel relaxation search algorithm. These algorithms were compared with several collaborative filtering approaches that do not consider the transitive associations: a simple graph search approach, two variations of the user-based approach, and an item-based approach. Our experimental results indicate that spreading activation-based approaches significantly outperformed the other collaborative filtering methods as measured by recommendation precision, recall, the F-measure, and the rank score.We also observed the over-activation effect of the spreading activation approach, that is, incorporating transitive associations with past transactional data that is not sparse may “dilute” the data used to infer user preferences and lead to degradation in recommendation performance.
Type:
Journal Article (Paginated)
Language:
en
Keywords:
Internet; World Wide Web; Informetrics
Local subject classification:
National Science Digital Library; NSDL; Artificial intelligence lab; AI lab; Collaborative filtering

Full metadata record

DC FieldValue Language
dc.contributor.authorHuang, Zanen_US
dc.contributor.authorChen, Hsinchunen_US
dc.contributor.authorZeng, Danielen_US
dc.date.accessioned2004-08-16T00:00:01Z-
dc.date.available2010-06-18T23:26:22Z-
dc.date.issued2004-01en_US
dc.date.submitted2004-08-16en_US
dc.identifier.citationApplying Associative Retrieval Techniques to Alleviate the Sparsity Problem in Collaborative Filtering 2004-01, 22(1):116-142 ACM Transactions on Information Systemsen_US
dc.identifier.urihttp://hdl.handle.net/10150/105493-
dc.descriptionArtificial Intelligence Lab, Department of MIS, University of Arizonaen_US
dc.description.abstractRecommender systems are being widely applied in many application settings to suggest products, services, and information items to potential consumers. Collaborative filtering, the most successful recommendation approach, makes recommendations based on past transactions and feedback from consumers sharing similar interests. A major problem limiting the usefulness of collaborative filtering is the sparsity problem, which refers to a situation in which transactional or feedback data is sparse and insufficient to identify similarities in consumer interests. In this article, we propose to deal with this sparsity problem by applying an associative retrieval framework and related spreading activation algorithms to explore transitive associations among consumers through their past transactions and feedback. Such transitive associations are a valuable source of information to help infer consumer interests and can be explored to deal with the sparsity problem. To evaluate the effectiveness of our approach, we have conducted an experimental study using a data set from an online bookstore. We experimented with three spreading activation algorithms including a constrained Leaky Capacitor algorithm, a branch-and-bound serial symbolic search algorithm, and a Hopfield net parallel relaxation search algorithm. These algorithms were compared with several collaborative filtering approaches that do not consider the transitive associations: a simple graph search approach, two variations of the user-based approach, and an item-based approach. Our experimental results indicate that spreading activation-based approaches significantly outperformed the other collaborative filtering methods as measured by recommendation precision, recall, the F-measure, and the rank score.We also observed the over-activation effect of the spreading activation approach, that is, incorporating transitive associations with past transactional data that is not sparse may “dilute” the data used to infer user preferences and lead to degradation in recommendation performance.en_US
dc.format.mimetypeapplication/pdfen_US
dc.language.isoenen_US
dc.publisherACMen_US
dc.subjectInterneten_US
dc.subjectWorld Wide Weben_US
dc.subjectInformetricsen_US
dc.subject.otherNational Science Digital Libraryen_US
dc.subject.otherNSDLen_US
dc.subject.otherArtificial intelligence laben_US
dc.subject.otherAI laben_US
dc.subject.otherCollaborative filteringen_US
dc.titleApplying Associative Retrieval Techniques to Alleviate the Sparsity Problem in Collaborative Filteringen_US
dc.typeJournal Article (Paginated)en_US
dc.identifier.journalACM Transactions on Information Systemsen_US
All Items in UA Campus Repository are protected by copyright, with all rights reserved, unless otherwise indicated.