A-Weighted-Method-to-Update-Network-User-Preference-Profile-Dynamically.pdf

52 International Journal of Interdisciplinary Telecommunications and Networking, 6(2), 52-67, April-June 2014 Copyright ? 2014, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. ABSTRACT There are many alternatives in a Recommender System (RS) that can be represented by numerical attributes. One of the most challenging tasks in developing RS is the design of techniques that can infer user preferences through observation of their actions. A RS usually stores a personal preference profile associated with each user, but the initial profile of a user is usually incomplete and imprecise. Therefore, it is necessary to update a user’s preference profile dynamically. Some previous research has covered this area, but neglected an important fact in real situations, where different weights should be considered for every attribute when selecting alternatives and updating a user preference profile. This paper provides a realis- tic and weighted method to update network user preferences through analysis of user selections. More specifically, an algorithm to compute and update weights of different attributes in a dynamic way is presented. The weights are used in the adaptation process of network user preference profile. The method is tested by extensive simulations and the obtained results show that it is more effective than previous methods. A Weighted Method to Update Network User Preference Profile Dynamically Zhi-Yuan Zhang, School of Communication and Information Engineering, Beijing Jiaotong University, Beijing, China Yun Liu, School of Communication and Information Engineering, Beijing Jiaotong University, Beijing, China Qing-An Zeng, Department of Computer Systems Technology, North Carolina AT State University, Greensboro, NC, USA Keywords: Network User Preference Profile, Numerical Attribute, Recommender System, Weighted DOI: 10.4018/ijitn.2014040103 Copyright ? 2014, IGI Global.