A fuzzy clustering method of construction of ontology-based user profiles

0965-9978/$ see front matter 2008 Elsevier Ltd. A doi:10.1016/j.advengsoft.2008.10.006 * Corresponding author. Address: College of Com neering, Hohai University, China. E-mail address: [email protected] (L. H With the rapid development of personalized information retrieval, user profile plays an important role. In this paper, we propose a fuzzy clustering method of construction of ontology-based user profiles (FCOU). In the FCOU method, we employ fuzzy clustering techniques combined with optimization techniques to develop ontology-based user profiles. One key feature of FCOU is that it employs an augmented Lagrangian function to create fuzzy clustering model for the construction of user profiles. Another key feature of FCOU is that it employs the combination of FCM, PHR and simulated annealing to develop ontology-based user profiles. The method allows some information to belong to several user profiles simultaneously with different degrees of accuracy. 2008 Elsevier Ltd. All rights reserved.