A graph-based recommender system for digital library

Zan Huang, Wingyan Chung, Thian Huat Ong, Hsinchun Chen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

117 Citations (Scopus)

Abstract

Research shows that recommendations comprise a valuable service for users of a digital library [11]. While most existing recommender systems rely either on a content-based approach or a collaborative approach to make recommendations, there is potential to improve recommendation quality by using a combination of both approaches (a hybrid approaches). In this paper, we report how we tested the idea of using a graph-based recommender system that naturally combines the content-based and collaborative approaches. Due to the similarity between our problem and a concept retrieval task, a Hopfield net algorithm was used to exploit high-degree book-book, user-user and book-user associations. Sample hold-out testing and preliminary subject testing were conducted to evaluate the system, by which it was found that the system gained improvement with respect to both precision and recall by combining content-based and collaborative approaches. However, no significant improvement was observed by exploiting high-degree associations.

Original languageEnglish (US)
Title of host publicationProceedings of the ACM International Conference on Digital Libraries
EditorsG. Marchionini, W. Hersh
Pages65-73
Number of pages9
StatePublished - 2002
EventProceedings of the Second ACM/IEEE-CS Joint Conference on Digital Libraries - Portland, OR, United States
Duration: Jul 14 2002Jul 18 2002

Other

OtherProceedings of the Second ACM/IEEE-CS Joint Conference on Digital Libraries
CountryUnited States
CityPortland, OR
Period7/14/027/18/02

Fingerprint

Digital libraries
Recommender systems
Testing

Keywords

  • Chinese phrase extraction
  • Collaborative filtering
  • Content-based filtering
  • Graph-based model
  • Hopfield net algorithm
  • Mutual information algorithm
  • Recommender system

ASJC Scopus subject areas

  • Computer Science(all)
  • Social Sciences(all)

Cite this

Huang, Z., Chung, W., Ong, T. H., & Chen, H. (2002). A graph-based recommender system for digital library. In G. Marchionini, & W. Hersh (Eds.), Proceedings of the ACM International Conference on Digital Libraries (pp. 65-73)

A graph-based recommender system for digital library. / Huang, Zan; Chung, Wingyan; Ong, Thian Huat; Chen, Hsinchun.

Proceedings of the ACM International Conference on Digital Libraries. ed. / G. Marchionini; W. Hersh. 2002. p. 65-73.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Huang, Z, Chung, W, Ong, TH & Chen, H 2002, A graph-based recommender system for digital library. in G Marchionini & W Hersh (eds), Proceedings of the ACM International Conference on Digital Libraries. pp. 65-73, Proceedings of the Second ACM/IEEE-CS Joint Conference on Digital Libraries, Portland, OR, United States, 7/14/02.
Huang Z, Chung W, Ong TH, Chen H. A graph-based recommender system for digital library. In Marchionini G, Hersh W, editors, Proceedings of the ACM International Conference on Digital Libraries. 2002. p. 65-73
Huang, Zan ; Chung, Wingyan ; Ong, Thian Huat ; Chen, Hsinchun. / A graph-based recommender system for digital library. Proceedings of the ACM International Conference on Digital Libraries. editor / G. Marchionini ; W. Hersh. 2002. pp. 65-73
@inproceedings{51565cddb14e429c87230f25da57bfd4,
title = "A graph-based recommender system for digital library",
abstract = "Research shows that recommendations comprise a valuable service for users of a digital library [11]. While most existing recommender systems rely either on a content-based approach or a collaborative approach to make recommendations, there is potential to improve recommendation quality by using a combination of both approaches (a hybrid approaches). In this paper, we report how we tested the idea of using a graph-based recommender system that naturally combines the content-based and collaborative approaches. Due to the similarity between our problem and a concept retrieval task, a Hopfield net algorithm was used to exploit high-degree book-book, user-user and book-user associations. Sample hold-out testing and preliminary subject testing were conducted to evaluate the system, by which it was found that the system gained improvement with respect to both precision and recall by combining content-based and collaborative approaches. However, no significant improvement was observed by exploiting high-degree associations.",
keywords = "Chinese phrase extraction, Collaborative filtering, Content-based filtering, Graph-based model, Hopfield net algorithm, Mutual information algorithm, Recommender system",
author = "Zan Huang and Wingyan Chung and Ong, {Thian Huat} and Hsinchun Chen",
year = "2002",
language = "English (US)",
pages = "65--73",
editor = "G. Marchionini and W. Hersh",
booktitle = "Proceedings of the ACM International Conference on Digital Libraries",

}

TY - GEN

T1 - A graph-based recommender system for digital library

AU - Huang, Zan

AU - Chung, Wingyan

AU - Ong, Thian Huat

AU - Chen, Hsinchun

PY - 2002

Y1 - 2002

N2 - Research shows that recommendations comprise a valuable service for users of a digital library [11]. While most existing recommender systems rely either on a content-based approach or a collaborative approach to make recommendations, there is potential to improve recommendation quality by using a combination of both approaches (a hybrid approaches). In this paper, we report how we tested the idea of using a graph-based recommender system that naturally combines the content-based and collaborative approaches. Due to the similarity between our problem and a concept retrieval task, a Hopfield net algorithm was used to exploit high-degree book-book, user-user and book-user associations. Sample hold-out testing and preliminary subject testing were conducted to evaluate the system, by which it was found that the system gained improvement with respect to both precision and recall by combining content-based and collaborative approaches. However, no significant improvement was observed by exploiting high-degree associations.

AB - Research shows that recommendations comprise a valuable service for users of a digital library [11]. While most existing recommender systems rely either on a content-based approach or a collaborative approach to make recommendations, there is potential to improve recommendation quality by using a combination of both approaches (a hybrid approaches). In this paper, we report how we tested the idea of using a graph-based recommender system that naturally combines the content-based and collaborative approaches. Due to the similarity between our problem and a concept retrieval task, a Hopfield net algorithm was used to exploit high-degree book-book, user-user and book-user associations. Sample hold-out testing and preliminary subject testing were conducted to evaluate the system, by which it was found that the system gained improvement with respect to both precision and recall by combining content-based and collaborative approaches. However, no significant improvement was observed by exploiting high-degree associations.

KW - Chinese phrase extraction

KW - Collaborative filtering

KW - Content-based filtering

KW - Graph-based model

KW - Hopfield net algorithm

KW - Mutual information algorithm

KW - Recommender system

UR - http://www.scopus.com/inward/record.url?scp=0036992318&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036992318&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0036992318

SP - 65

EP - 73

BT - Proceedings of the ACM International Conference on Digital Libraries

A2 - Marchionini, G.

A2 - Hersh, W.

ER -