Designing a Web-based Testing Tool for Multi-Criteria Recommender Systems
Nikos Manouselis, Constantina I. Costopoulou · 2006
A plethora of real-life applications of recommender systems in the Web exists. These systems help users to deal with information overload, by providing personalized recommendations regarding online content and services. Due to the dynamic and changing parameters of the various application contexts, careful testing and parameterization has to be carried out before a recommender system is finally deployed in a real setting. This paper proposes a Web-based tool that allows for simulated testing of a special class of multi-criteria recommender systems, namely multi-attribute collaborative filtering systems. More specifically, it introduces a number of collaborative filtering algorithms that are based on Multi-Attribute Utility Theory (MAUT), and presents the design and implementation of a Web-based tool termed as the Collaborative Filtering Simulator (CollaFiS), which may be used for their simulated testing. The specification of CollaFiS is described using the Unified Modelling Language (UML). In addition, a characteristic scenario of the