Using web usage mining and SVD to improve E-commerce recommendation quality

Jae Kyeong Kim; Yoon Ho Cho

doi:10.1007/978-3-540-39896-7_8

Using web usage mining and SVD to improve E-commerce recommendation quality

Jae Kyeong Kim
, Yoon Ho Cho

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

14 Citations (Scopus)

Abstract

Collaborative filtering is the most successful recommendation method, but its widespread use has exposed some well-known limitations, such as sparsity and scalability. This paper proposes a recommendation methodology based on Web usage mining and SVD (Singular Value Decomposition) to enhance the recommendation quality and the system performance of current collaborative filtering-based recommender systems. Web usage mining populates the rating database by tracking customers' shopping behaviors on the Web, so leading to better quality recommendations. SVD is used to improve the performance of searching for nearest neighbors through dimensionality reduction of the rating database. Several experiments on real Web retailer data show that the proposed methodology provides higher quality recommendations and better performance than other recommendation methodologies.

Original language	English
Title of host publication	Intelligent Agents and Multi-Agent Systems
Editors	Jaeho Lee, Mike Barley
Publisher	Springer Verlag
Pages	86-97
Number of pages	12
ISBN (Electronic)	9783540204602
DOIs	https://doi.org/10.1007/978-3-540-39896-7_8
Publication status	Published - 2003
Event	6th Pacific Rim International Workshop on Multi-Agents, PRIMA 2003 - Seoul, Korea, Republic of Duration: 7 Nov 2003 → 8 Nov 2003

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	2891
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	6th Pacific Rim International Workshop on Multi-Agents, PRIMA 2003
Country/Territory	Korea, Republic of
City	Seoul
Period	7/11/03 → 8/11/03

Bibliographical note

Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 2003.

Access to Document

10.1007/978-3-540-39896-7_8

Cite this

Kim, J. K., & Cho, Y. H. (2003). Using web usage mining and SVD to improve E-commerce recommendation quality. In J. Lee, & M. Barley (Eds.), Intelligent Agents and Multi-Agent Systems (pp. 86-97). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2891). Springer Verlag. https://doi.org/10.1007/978-3-540-39896-7_8

@inproceedings{9a78b3022e6a4aefbcfaca0591b22d56,

title = "Using web usage mining and SVD to improve E-commerce recommendation quality",

abstract = "Collaborative filtering is the most successful recommendation method, but its widespread use has exposed some well-known limitations, such as sparsity and scalability. This paper proposes a recommendation methodology based on Web usage mining and SVD (Singular Value Decomposition) to enhance the recommendation quality and the system performance of current collaborative filtering-based recommender systems. Web usage mining populates the rating database by tracking customers' shopping behaviors on the Web, so leading to better quality recommendations. SVD is used to improve the performance of searching for nearest neighbors through dimensionality reduction of the rating database. Several experiments on real Web retailer data show that the proposed methodology provides higher quality recommendations and better performance than other recommendation methodologies.",

author = "Kim, \{Jae Kyeong\} and Cho, \{Yoon Ho\}",

note = "Publisher Copyright: {\textcopyright} Springer-Verlag Berlin Heidelberg 2003.; 6th Pacific Rim International Workshop on Multi-Agents, PRIMA 2003 ; Conference date: 07-11-2003 Through 08-11-2003",

year = "2003",

doi = "10.1007/978-3-540-39896-7\_8",

language = "English",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "86--97",

editor = "Jaeho Lee and Mike Barley",

booktitle = "Intelligent Agents and Multi-Agent Systems",

address = "Germany",

}

Kim, JK & Cho, YH 2003, Using web usage mining and SVD to improve E-commerce recommendation quality. in J Lee & M Barley (eds), Intelligent Agents and Multi-Agent Systems. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2891, Springer Verlag, pp. 86-97, 6th Pacific Rim International Workshop on Multi-Agents, PRIMA 2003, Seoul, Korea, Republic of, 7/11/03. https://doi.org/10.1007/978-3-540-39896-7_8

Using web usage mining and SVD to improve E-commerce recommendation quality. / Kim, Jae Kyeong; Cho, Yoon Ho.
Intelligent Agents and Multi-Agent Systems. ed. / Jaeho Lee; Mike Barley. Springer Verlag, 2003. p. 86-97 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2891).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Using web usage mining and SVD to improve E-commerce recommendation quality

AU - Kim, Jae Kyeong

AU - Cho, Yoon Ho

PY - 2003

Y1 - 2003

N2 - Collaborative filtering is the most successful recommendation method, but its widespread use has exposed some well-known limitations, such as sparsity and scalability. This paper proposes a recommendation methodology based on Web usage mining and SVD (Singular Value Decomposition) to enhance the recommendation quality and the system performance of current collaborative filtering-based recommender systems. Web usage mining populates the rating database by tracking customers' shopping behaviors on the Web, so leading to better quality recommendations. SVD is used to improve the performance of searching for nearest neighbors through dimensionality reduction of the rating database. Several experiments on real Web retailer data show that the proposed methodology provides higher quality recommendations and better performance than other recommendation methodologies.

AB - Collaborative filtering is the most successful recommendation method, but its widespread use has exposed some well-known limitations, such as sparsity and scalability. This paper proposes a recommendation methodology based on Web usage mining and SVD (Singular Value Decomposition) to enhance the recommendation quality and the system performance of current collaborative filtering-based recommender systems. Web usage mining populates the rating database by tracking customers' shopping behaviors on the Web, so leading to better quality recommendations. SVD is used to improve the performance of searching for nearest neighbors through dimensionality reduction of the rating database. Several experiments on real Web retailer data show that the proposed methodology provides higher quality recommendations and better performance than other recommendation methodologies.

UR - https://www.scopus.com/pages/publications/0344823863

U2 - 10.1007/978-3-540-39896-7_8

DO - 10.1007/978-3-540-39896-7_8

M3 - Conference contribution

AN - SCOPUS:0344823863

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 86

EP - 97

BT - Intelligent Agents and Multi-Agent Systems

A2 - Lee, Jaeho

A2 - Barley, Mike

PB - Springer Verlag

T2 - 6th Pacific Rim International Workshop on Multi-Agents, PRIMA 2003

Y2 - 7 November 2003 through 8 November 2003

ER -

Using web usage mining and SVD to improve E-commerce recommendation quality

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this