Authors: Hyung Jun Ahn, Jong Woo Kim
Addresses: Department of Management Systems, Waikato Management School, University of Waikato, Private Bag 3105, Hamilton, New Zealand. ' School of Business, Hanyang University, Seoul, 133-791, Korea
Abstract: One of the widely used methods for product recommendation in internet storefronts is matching product features with target customer profiles. When using this method, it is very important to choose a suitable subset of features for recommendation efficiency and performance, which, however, has not been rigorously researched so far. In this paper, we utilise a dataset collected from a virtual shopping experiment in a Korean internet book shopping mall to compare several popular methods of feature selection from other disciplines for product recommendation: the vector-space model, Term Frequency-Inverse Document Frequency (TFIDF), the Mutual Information (MI) method and the Singular Value Decomposition (SVD). The application of SVD showed the best performance in the analysis results.
Keywords: product recommendation; feature selection; content-based filtering; singular value decomposition; SVD; electronic business; internet shopping; online shopping; e-business; electronic business; product features; customer profiles; virtual shopping; Korea; book shopping.
International Journal of Electronic Business, 2006 Vol.4 No.5, pp.432 - 444
Published online: 17 Nov 2006 *Full-text access for editors Access for subscribers Purchase this article Comment on this article