[ Previous [ 1] [ 2] [ 3] [ 4] [ 5] [ 6] [ 7] [ 8] [ 9] [ 10] [ 11] [ 12] [ 13] [ 14] [ 15] [ 16] [ 17] [ 18] [ 19] [ 20]


Journal of Information Science and Engineering, Vol. 24 No. 1, pp. 129-142 (January 2008)

A Data Mining-Based Method for the Incremental Update of Supporting Personalized Information Filtering*

Ye-In Chang, Jun-Hong Shen and Tsu-I Chen
Department of Computer Science and Engineering
National Sun Yat-Sen University
Kaohsiung, 804 Taiwan
E-mail: changyi@cse.nsysu.edu.tw

Information filtering is an area of research that develops tools for discriminating between relevant and irrelevant information. Users first give descriptions about what they need, i.e., user profiles represented by a set of keywords, to start the services. A profile index is built on these profiles. Then, the Web page will be recommended to the users whose profiles belong to the filtered results. Therefore, a critical issue of the information filtering service is how to index the user profiles for an efficient matching process. Among previous proposed methods, Wu and Chens graph-based index method can expect to minimize the storage space. However, when the users often change their interests, the index structure of Wu and Chens method needs to be reconstructed, resulting in the high update cost. Therefore, in this paper, we propose a data mining-based method for the incremental update of the index structure, the updatable tree, to reduce the update cost. In fact, each keyword could have a weight representing the degree of importance to a user. We apply this feature to distinguish between long-term and short-term interests. By making use of the property that the short-term interest has a higher probability to be changed than the long-term one, our proposed method can locally update the short-term interest, resulting in the low update cost. According to our experimental results, our method really can reduce the update cost as needed by Wu and Chens method.

Keywords: data mining, incremental update, information filtering, personalization, profile

Full Text () Retrieve PDF document (200801_09.pdf)

Received February 2, 2007; accepted July 13, 2007.
Communicated by K. Robert Lai, Yu-Chee Tseng and Shu-Yuan Chen.
*This research was supported in part by the National Science Council of Taiwan, R.O.C. under grant No. NSC 95-2221-E-110-101 and by National Sun Yat-Sen University.