International Conference on Information, Electronic and Computer Science (ICIECS 2010 E-BOOK)

Zibo,China,11.26-11.28,2010

ISBN: 978-1-935068-42-6 Scientific Research Publishing, USA

E-Book 2224pp Pub. Date: November 2010

Category: Computer Science & Communications

Price: $360

Title: A Recommendation System Model Based on Web Mining
Source: International Conference on Information, Electronic and Computer Science (ICIECS 2010 E-BOOK) (pp 682-687)
Author(s): Ting Zhang, Department. Of computer science and technology of Tangshan College,Heibei,Tangshan,China
Shucai Fu, Department. Of computer science and technology of Tangshan College,Heibei,Tangshan,China
Abstract: Traditional collaborative filtering recommendation is difficult to provide high quality recommendation for non-registered users. In this paper, express method of URL-UserID associated matrix for browsing behaviors of web users was described, and model of web page and web consumer clustering was established. A recommendation system model based on web clustering was proposed. It analyzed web usage data, web content data and web structure data by clustering and recommendation engine provided recommendation service according to results of data mining. In end of this paper accuracy of model was tested. Experimental results showed that, web page and web consumer clustering based on URL-UserID associated matrix was an effective clustering method, it provided scientific basis for e-commerce and some web-based information services, such as marketing, contacting customers, discovering potential customers and providing personalized services, etc. Quality of recommendation system based on web clustering was better than quality of traditional collaborative filtering recommendation system.
Free SCIRP Newsletters
Copyright © 2006-2024 Scientific Research Publishing Inc. All Rights Reserved.
Top