Affiliations: School of Information Sciences, University of Pittsburgh, 135 N. Bellefield Ave., Pittsburgh, PA 15260, USA
Note:  This work was partially supported by NSF CRCD/EI Award 0426021 and NSF CAREER Award 0447083.
Abstract: Adaptive Web systems utilize user models (or group models) to represent essential information about an individual user (or a group) so that these systems can adapt their behavior to the goals, tasks, interests, and other features of individual users or groups of users. Meanwhile, proper performance assessment has become a critical issue for the efficient deployment of adaptive Web systems that implement complex user model inferences. This paper presents one of the first efforts to develop a framework for evaluating the performance of user modeling servers (UMS). We conduct a performance-driven analysis of the UMS conceptual model, extracting a comprehensive set of parameters in order to build a practical UMS Performance Evaluation Framework (UMS/PEF). We also apply the proposed UMS/PEF framework for comparing performances between different UMS. Experimental results have demonstrated high utility of the proposed UMS/PEF framework.
Keywords: Adaptive web systems, performance evaluation, user modeling server