AI帮你理解科学

AI 生成解读视频

AI抽取解析论文重点内容自动生成视频


pub
生成解读视频

AI 溯源

AI解析本论文相关学术脉络


Master Reading Tree
生成 溯源树

AI 精读

AI抽取本论文的概要总结


微博一下
We have proposed an intelligent web recommender system known as SWARS based on sequential web access patterns

An intelligent recommender system using sequential Web access patterns

(2004)

被引用83|浏览7
下载 PDF 全文
引用
微博一下

摘要

To provide intelligent personalized online services such as web recommendations, it is usually necessary to model users’ web access behavior. To achieve this, one of the promising approaches is web usage mining, which mines web logs for user models and recommendations. Different from most web recommender systems that are mainly based on c...更多

代码

数据

0
简介
  • With the explosive growth of information available on the World Wide Web, it has become much more difficult to access relevant information from the Web.
  • This paper proposes a compact data model, called Pattern-tree, which stores the sequential web access patterns, and an efficient approach for user pattern matching and recommendation rules generation.
  • A Pattern-tree model is proposed for storing sequential web access patterns compactly, so that it can be used for matching with a user’s current access sequence and generating recommendation rules more efficiently in the Recommendation Rules Generation component.
重点内容
  • With the explosive growth of information available on the World Wide Web, it has become much more difficult to access relevant information from the Web
  • Different from the majority of the existing web recommendation techniques, we propose an intelligent web recommender system known as SWARS (Sequential Web Access-based Recommender System) that uses a sequential pattern mining technique
  • A Pattern-tree model is proposed for storing sequential web access patterns compactly, so that it can be used for matching with a user’s current access sequence and generating recommendation rules more efficiently in the Recommendation Rules Generation component
  • We can conclude that better recommendations can be obtained with smaller support thresholds, at the expense of increased computational complexity for sequential web access pattern mining and maintaining a larger Patterntree
  • We have proposed an intelligent web recommender system known as SWARS based on sequential web access patterns
  • The mined patterns are stored in the Pattern-tree, which is used for matching and generating web links for online recommendations
结果
  • The matching path will not exist when the length of the current access sequence is longer than the depth of the Pattern-tree.
  • Some initial items can be removed to make the current access sequence shorter than the depth of the Pattern-tree before the sequence matching process begins.
  • In order to improve the precision of recommendation rules generation, only web access sequence whose length is not less than a given threshold can be processed.
  • The recommendation rules for the current access sequence S = cab are generated as {a, c}, which are ordered by their support.
  • Complexity analysis: The cost of looking up the current access sequence S of length m in a Pattern-tree T with MinLength = Lmin and MaxLength = Lmax is O(min(m, Lmax )).
  • For the prefix sequence Sprefix = a1a2...ak (k ≥ MinLength), the authors generate a recommendation rule RR = {e1, e2, ..., em} using the Pattern-tree, where all events are ordered by their support.
  • As the Pattern-tree only stores web access sub-sequences accessed frequently by users, the recommendation rules generation approach is unable to find recommended pages if the current access sequence does not include a frequent suffix sequence, in which case the generated recommendation rule is empty.
  • The three main components of the proposed SWARS system, Sequential Pattern Mining, Pattern-tree Construction and Recommendation Rules Generation, were implemented in C++.
  • The authors measured the scalability of the Sequential Pattern Mining and Pattern-tree Construction processes with respect to different support thresholds.
  • The scalability of the precision, satisfaction, and applicability measures of recommendation rules generation had been measured with respect to different support thresholds.
结论
  • The authors can conclude that better recommendations can be obtained with smaller support thresholds, at the expense of increased computational complexity for sequential web access pattern mining and maintaining a larger Patterntree.
  • The last experiment measured the scalability of the satisfaction of recommendation rules generation with respect to different numbers of steps.
  • The proposed system has achieved good performance with high satisfaction and applicability
表格
  • Table1: A SAMPLE DATABASE OF WEB ACCESS SEQUENCES
  • Table2: SEQUENTIAL WEB ACCESS PATTERNS WITH MINSUP = 3 FROM THE SAMPLE DATABASE IN TABLE 1
Download tables as Excel
引用论文
  • M. Eirinaki and M. Vazirgiannis, “Web mining for web personalization”, ACM Transactions on Internet Technology, Vol. 3, No. 1, 2003, pp. 127.
    Google ScholarLocate open access versionFindings
  • J. Konstan, B. Miller, D. Maltz, J. Herlocker, L. Gordon, and J. Riedl, “GroupLens: applying collaborative filtering to usenet news”, Communications of the ACM, 40(3), 1997, pp. 77-87.
    Google ScholarLocate open access versionFindings
  • T.W. Yan and H. Garcia-Molina, “The SIFT information dissemination system”, ACM Transactions on Database Systems, Vol. 24, No. 4, 1999, pp.529-565.
    Google ScholarLocate open access versionFindings
  • T. Joachims, D. Freitag, and T. Mitchel, “WebWatcher: a tour guide for the World Wide Web”, Proc. of the 5th International Joint Conference on AI, Japan, 1997, pp. 770-775.
    Google ScholarLocate open access versionFindings
  • C. Shahabi, F. Banaei-Kashani, Y. Chen, and D. McLeod, “Yoda: an accurate and scalable web-based recommendation system”, Proc. of the 6th International Conference on Cooperative Information Systems (CoopIS 2001), Trento, Italy, September 2001.
    Google ScholarLocate open access versionFindings
  • B. Mobasher, H. Dai, T. Luo and M. Nakagawa, “Effective personalization based on association rule discovery from web usage data”, Proc. of the 3rd ACM Workshop on Web Information and Data Management (WIDM01), 2001.
    Google ScholarLocate open access versionFindings
  • W. Lin, S. A.Alvarez, and C. Ruiz, “Collaborative recommendation via adaptive association rule mining”, Proc. of the Web Mining for ECommerce Workshop (WebKDD'2000), August 2000, Boston.
    Google ScholarLocate open access versionFindings
  • B. Mobasher, “A web personalization engine based on user transaction clustering”, Proc. of the 9th Workshop on Information Technologies and Systems (WITS'99), December 1999.
    Google ScholarLocate open access versionFindings
  • D.S. Phatak, and R. Mulvaney, “Clustering for personalized mobile web usage”, Proc. of the IEEE FUZZ'02, Hawaii, May 2002, pp. 705-710.
    Google ScholarLocate open access versionFindings
  • R. Agrawal, and R. Srikant, “Mining Sequential Patterns”, Proc. of the 11th International Conference on Data Engineering, Taiwan, 1995.
    Google ScholarLocate open access versionFindings
  • B.Y. Zhou, S.C. Hui, and A.C.M. Fong, “CS-mine: an efficient WAPtree mining for web access patterns”, the 6th Asia Pacific Web Conference (APWEB04), Hangzhou, China, April 14-17, 2004, pp.523532.
    Google ScholarLocate open access versionFindings
  • R. Srikant, and R. Agrawal, “Mining sequential patterns: generalizations and performance improvements”, Proc. of the 5th International Conference on Extending Database Technology (EDBT), Avignon, France, 1996, pp. 3-17.
    Google ScholarLocate open access versionFindings
  • R. Cooley, B. Mobasher, and J. Srivastava, “Data preparation for mining World Wide Web browsing patterns”, Journal of Knowledge and Information Systems, Vol. 1, No. 1. 1999.
    Google ScholarLocate open access versionFindings
  • D. Knuth, “The art of computer programming”, Vol. 2, 2nd edition, Addison-Wesley, October 1998.
    Google ScholarFindings
作者
bing zhou
bing zhou
koling chang
koling chang
您的评分 :
0

 

标签
评论
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn
小科