A web recommender system based on dynamic sampling of user information access behaviors

From WikiPapers
Jump to: navigation, search

A web recommender system based on dynamic sampling of user information access behaviors is a 2009 conference paper written in English by Chen J., Shtykh R.Y., Jin Q. and published in Proceedings - IEEE 9th International Conference on Computer and Information Technology, CIT 2009.

[edit] Abstract

In this study, we propose a Gradual Adaption Model for a Web recommender system. This model is used to track users' focus of interests and its transition by analyzing their information access behaviors, and recommend appropriate information. A set of concept classes are extracted from Wikipedia. The pages accessed by users are classified by the concept classes, and grouped into three terms of short, medium and long periods, and two categories of remarkable and exceptional for each concept class, which are used to describe users' focus of interests, and to establish reuse probability of each concept class in each term for each user by Full Bayesian Estimation as well. According to the reuse probability and period, the information that a user is likely to be interested in is recommended. In this paper, we propose a new approach by which short and medium periods are determined based on dynamic sampling of user information access behaviors. We further present experimental simulation results, and show the validity and effectiveness of the proposed system.

[edit] References

This section requires expansion. Please, help!

Cited by

Probably, this publication is cited by others, but there are no articles available for them in WikiPapers.