Small Clues Tell: a Collaborative Expansion Approach for Effective Content-Based Recommendations

ABSTRACT

Content-based recommendation techniques usually require a large number of training examples for model construction, which however may not always be available in many real-world scenarios. To address the training data availability constraint common to the content-based approach, we develop a collaborative expansion-based approach to expand the size of training examples, which could lead to improved content-based recommendations. We use a book rating data set collected from Amazon to evaluate our proposed method and compare its performance against those of two salient benchmark techniques. The results show that our method outperforms the benchmark techniques consistently and significantly. Our method expands the size of training examples for a focal customer by leveraging the available preferences of his or her referent group, and thereby better supports personalized recommendations than existing techniques that solely follow content-based or collaborative filtering, without incurring costs to identify, collect, and analyze additional information. This study reveals the value and feasibility of collaborative expansion as a viable means to increase training size for the focal customer and thus address the training data availability constraint that seriously hinders the performance of content-based recommender systems.

KEYWORDS:

• Recommender systems
• content-based recommendations
• data availability constraint
• training set expansion
• small-size training set problem

Notes

¹ We obtained this sample of book ratings from http://www.amazon.com.

Additional information

Funding

This work was partially supported by the [Ministry of Science and Technology, Taiwan] under Grants [108-2218-E-009-051, 107-2410-H-415-011-MY3, and 106-2410-H-002-068-MY3].

Notes on contributors

Yen-Hsien Lee

Yen-Hsien Lee received his BS, MBA, and Ph.D. in Information Management from National SunYat-Sen University in Taiwan in 1998, 2000, and 2005, respectively. He is currently anassociate professor of Department of Management Information Systems at National ChiayiUniversity in Taiwan. He was a visiting scholar at University of Utah in Fall 2002 and at University of Florida in Fall 2016. His papers have appeared in Journal of ManagementInformation Systems, ACM Transactions on Management Information Systems, Artificial Intelligence in Medicine, Journal of the American Society for Information Science and Technology, IEEE Transactions on Systems, Man and Cybernetics, Decision Support Systems, and Expert Systems with Applications. His current research interests include knowledge discovery and data mining, knowledge management, information retrieval, text mining, andweb mining.

Chih-Ping Wei

Chih-Ping Wei received a BS in Management Science from the National Chiao-Tung Universityin Taiwan and an MS and a Ph.D. in Management Information Systems from the University ofArizona. He is currently a professor of Department of Information Management at NationalTaiwan University, Taiwan. His papers have appeared in Journal of Management InformationSystems, European Journal of Information Systems, Decision Sciences, Decision SupportSystems, IEEE Transactions on Engineering Management, IEEE Software, IEEE IntelligentSystems, IEEE Transactions on Systems, Man, and Cybernetics, IEEE Transactions onInformation Technology in Biomedicine, Journal of the American Society for InformationScience and Technology, etc. His current research interests include data analytics and businessintelligence, text mining and information retrieval, patent analysis and mining, and healthinformatics. He has edited special issues for Decision Support Systems, International Journalof Electronic Commerce, Electronic Commerce Research and Applications, InformationProcessing and Management, Pacific Asia Journal of the Association for Information Systems,etc.

Paul Jen-Hwa Hu

Paul Jen-Hwa Hu is David Eccles Chair Professor at the David Eccles School of Business, theUniversity of Utah. He has a Ph.D. in Management Information Systems from the Universityof Arizona. His current research interests include information technology in health care,technology implementation and management, business analytics, digital transformation,technology-empowered learning, and knowledge and innovation management. Hu haspublished papers in Journal of Organizational Computing and Electronic Commerce,Management Information Systems Quarterly, Information Systems Research, Journal ofManagement Information Systems, Journal of the Association for Information Systems,Decision Sciences, Decision Support Systems, Journal of Information Systems, Journal ofService Research, Journal of Business Research, Journal of the American Society forInformation Science and Technology, Journal of Biomedical informatics, and various IEEEand ACM journals and transactions.

Tsang-Hsiang Cheng

Tsang-Hsiang Cheng Cheng received a BS in Computer Science and an MBA in ManagementInformation Systems from the National Chiao-Tung University in Taiwan, and a Ph.D. inManagement Information Systems from the National Sun Yat-sen University in Taiwan. He iscurrently a professor of Department of Business Administration at Southern Taiwan Universityof Science and Technology in Taiwan. His papers have appeared in Journal of ManagementInformation Systems, IEEE Transactions on Systems, Man, and Cybernetics, Decision SupportSystems, Information Processing and Management, Journal of Database Management, etc.His current research interests include knowledge discovery and data mining, informationretrieval and text mining, knowledge management, and medical informatics.

Ci-Wei Lan

Ci-Wei Lan is an information architect at Global Business Solution Center, IBM (Taiwan). Hereceived Ph.D. degree from National Central University, Taiwan. His research interests areService-Oriented Computing and Big Data Technologies. He has practical experience ofinformation modeling and analytics in healthcare and retail domains. He is the core oforganizing member of IEEE ICEBE conference and SOAIC workshop for advocating serviceorientedtechnologies and e-business applications.