A clickstream-based collaborative filtering personalization model: Towards a better performance

Dong Ho Kim, Il Im, Nabil Adam, Vijayalakshmi Atluri, Michael Bieber, Yelena Yesha

Research output: Contribution to conferencePaper

38 Citations (Scopus)

Abstract

In recent years, clickstream-based Web personalization models for collaborative filtering recommendation have received much attention mainly due to their scalability [10,16,19]. The common personalization models are the Markov model, (sequential) association rule, and clustering. These models have shown strengths and weaknesses in their performance: for instance, the Markov model has higher precision and lower recall than (sequential) association rule and clustering, and vice versa [22]. In order to address the trade-off relationship of precision and recall, some study has combined two or more different models [22] or applied multi-order models [24,27]. The performance increases by these models, however, are at best marginal and still there is room for improving the performance because of their first order (one model type) application in making recommendation. We propose a new hybrid model for improving the performance, especially recall. The proposed hybrid model applies four prediction models - the Markov model, sequential association rule, association rule, and a default model [1,17] - in tandem in their precision order. We evaluated our model with Web usage data, and the result is promising.

Original languageEnglish
Pages88-95
Number of pages8
Publication statusPublished - 2004 Dec 1
EventWIDM 2004: Proceedings of the Sixth ACM International Workshop on Web Information and Data Management - Washington, DC, United States
Duration: 2004 Nov 122004 Nov 13

Other

OtherWIDM 2004: Proceedings of the Sixth ACM International Workshop on Web Information and Data Management
CountryUnited States
CityWashington, DC
Period04/11/1204/11/13

Fingerprint

Collaborative filtering
Association rules

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Information Systems

Cite this

Kim, D. H., Im, I., Adam, N., Atluri, V., Bieber, M., & Yesha, Y. (2004). A clickstream-based collaborative filtering personalization model: Towards a better performance. 88-95. Paper presented at WIDM 2004: Proceedings of the Sixth ACM International Workshop on Web Information and Data Management, Washington, DC, United States.
Kim, Dong Ho ; Im, Il ; Adam, Nabil ; Atluri, Vijayalakshmi ; Bieber, Michael ; Yesha, Yelena. / A clickstream-based collaborative filtering personalization model : Towards a better performance. Paper presented at WIDM 2004: Proceedings of the Sixth ACM International Workshop on Web Information and Data Management, Washington, DC, United States.8 p.
@conference{ca1ef5e1f7e449cebb05ecedecb46095,
title = "A clickstream-based collaborative filtering personalization model: Towards a better performance",
abstract = "In recent years, clickstream-based Web personalization models for collaborative filtering recommendation have received much attention mainly due to their scalability [10,16,19]. The common personalization models are the Markov model, (sequential) association rule, and clustering. These models have shown strengths and weaknesses in their performance: for instance, the Markov model has higher precision and lower recall than (sequential) association rule and clustering, and vice versa [22]. In order to address the trade-off relationship of precision and recall, some study has combined two or more different models [22] or applied multi-order models [24,27]. The performance increases by these models, however, are at best marginal and still there is room for improving the performance because of their first order (one model type) application in making recommendation. We propose a new hybrid model for improving the performance, especially recall. The proposed hybrid model applies four prediction models - the Markov model, sequential association rule, association rule, and a default model [1,17] - in tandem in their precision order. We evaluated our model with Web usage data, and the result is promising.",
author = "Kim, {Dong Ho} and Il Im and Nabil Adam and Vijayalakshmi Atluri and Michael Bieber and Yelena Yesha",
year = "2004",
month = "12",
day = "1",
language = "English",
pages = "88--95",
note = "WIDM 2004: Proceedings of the Sixth ACM International Workshop on Web Information and Data Management ; Conference date: 12-11-2004 Through 13-11-2004",

}

Kim, DH, Im, I, Adam, N, Atluri, V, Bieber, M & Yesha, Y 2004, 'A clickstream-based collaborative filtering personalization model: Towards a better performance' Paper presented at WIDM 2004: Proceedings of the Sixth ACM International Workshop on Web Information and Data Management, Washington, DC, United States, 04/11/12 - 04/11/13, pp. 88-95.

A clickstream-based collaborative filtering personalization model : Towards a better performance. / Kim, Dong Ho; Im, Il; Adam, Nabil; Atluri, Vijayalakshmi; Bieber, Michael; Yesha, Yelena.

2004. 88-95 Paper presented at WIDM 2004: Proceedings of the Sixth ACM International Workshop on Web Information and Data Management, Washington, DC, United States.

Research output: Contribution to conferencePaper

TY - CONF

T1 - A clickstream-based collaborative filtering personalization model

T2 - Towards a better performance

AU - Kim, Dong Ho

AU - Im, Il

AU - Adam, Nabil

AU - Atluri, Vijayalakshmi

AU - Bieber, Michael

AU - Yesha, Yelena

PY - 2004/12/1

Y1 - 2004/12/1

N2 - In recent years, clickstream-based Web personalization models for collaborative filtering recommendation have received much attention mainly due to their scalability [10,16,19]. The common personalization models are the Markov model, (sequential) association rule, and clustering. These models have shown strengths and weaknesses in their performance: for instance, the Markov model has higher precision and lower recall than (sequential) association rule and clustering, and vice versa [22]. In order to address the trade-off relationship of precision and recall, some study has combined two or more different models [22] or applied multi-order models [24,27]. The performance increases by these models, however, are at best marginal and still there is room for improving the performance because of their first order (one model type) application in making recommendation. We propose a new hybrid model for improving the performance, especially recall. The proposed hybrid model applies four prediction models - the Markov model, sequential association rule, association rule, and a default model [1,17] - in tandem in their precision order. We evaluated our model with Web usage data, and the result is promising.

AB - In recent years, clickstream-based Web personalization models for collaborative filtering recommendation have received much attention mainly due to their scalability [10,16,19]. The common personalization models are the Markov model, (sequential) association rule, and clustering. These models have shown strengths and weaknesses in their performance: for instance, the Markov model has higher precision and lower recall than (sequential) association rule and clustering, and vice versa [22]. In order to address the trade-off relationship of precision and recall, some study has combined two or more different models [22] or applied multi-order models [24,27]. The performance increases by these models, however, are at best marginal and still there is room for improving the performance because of their first order (one model type) application in making recommendation. We propose a new hybrid model for improving the performance, especially recall. The proposed hybrid model applies four prediction models - the Markov model, sequential association rule, association rule, and a default model [1,17] - in tandem in their precision order. We evaluated our model with Web usage data, and the result is promising.

UR - http://www.scopus.com/inward/record.url?scp=18744391091&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=18744391091&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:18744391091

SP - 88

EP - 95

ER -

Kim DH, Im I, Adam N, Atluri V, Bieber M, Yesha Y. A clickstream-based collaborative filtering personalization model: Towards a better performance. 2004. Paper presented at WIDM 2004: Proceedings of the Sixth ACM International Workshop on Web Information and Data Management, Washington, DC, United States.