marService: multiattribute utility recommendation for e-markets Online publication date: Wed, 10-Dec-2008
by Nikos Manouselis, Constantina Costopoulou
International Journal of Computer Applications in Technology (IJCAT), Vol. 33, No. 2/3, 2008
Abstract: Recommender systems are deployed in electronic commerce (e-commerce) settings to help customers find products according to their preferences. Product recommendations may help buyers to save time by helping them choose from a variety of options. Recommendations that take into account the multiple attributes affecting a potential buyer's decision can be particularly useful in the context of Business-to-Consumer (B2C) electronic markets (e-markets). Nevertheless, multiattribute recommender systems are usually more sophisticated than single-attribute ones, and their implementation may prove complex to e-market system developers. This paper presents the design, development and evaluation of marService, a product recommendation service that is based on Multi-Attribute Utility Theory (MAUT). This approach studies the application of marService for providing wine recommendations in an existing e-market and presents the results of a simulation experiment. Using an appropriate simulation environment, the evaluation of several design options for a set of algorithms for multiattribute utility recommendation has taken place, on two synthetic data sets for wine evaluations. Based on the experience from this experiment, some general suggestions that may prove useful to e-market developers wishing to implement a marService are also provided.
Online publication date: Wed, 10-Dec-2008
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Computer Applications in Technology (IJCAT):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email email@example.com