Tag-based smoothing for item recommendation

Jing Peng, Dajun Zeng

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

3 Citations (Scopus)

Abstract

Collaborative tagging sites allow users to save and annotate their favorite Web contents with tags. These tags provide a novel source of information for collaborative filtering. However, major gap exists in how to integrate tagging information into traditional recommender systems for better recommendation quality, due to the difficulty to quantize these semantic tags. This paper proposes a novel approach to convert this semantic information into quantitative values from a smoothing point of view, taking advantage of the topicbased method, and then make recommendation in a traditional user-based CF fashion based on the smoothed user-item matrix. Experiments on two real-world collaborative tagging datasets prove the effectiveness of our approach.

Original languageEnglish (US)
Title of host publicationProceedings of 2010 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2010
Pages452-456
Number of pages5
DOIs
StatePublished - 2010
Event2010 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2010 - QingDao, China
Duration: Jul 15 2010Jul 17 2010

Other

Other2010 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2010
CountryChina
CityQingDao
Period7/15/107/17/10

Fingerprint

Tag
Smoothing
Collaborative tagging
Collaborative filtering
Sources of information
Experiment
Recommender systems
Tagging
World Wide Web

Keywords

  • Collaborative filtering
  • Recommender systems
  • Smoothing
  • Tag
  • Weighing

ASJC Scopus subject areas

  • Information Systems and Management
  • Management Science and Operations Research

Cite this

Peng, J., & Zeng, D. (2010). Tag-based smoothing for item recommendation. In Proceedings of 2010 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2010 (pp. 452-456). [5551603] https://doi.org/10.1109/SOLI.2010.5551603

Tag-based smoothing for item recommendation. / Peng, Jing; Zeng, Dajun.

Proceedings of 2010 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2010. 2010. p. 452-456 5551603.

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

Peng, J & Zeng, D 2010, Tag-based smoothing for item recommendation. in Proceedings of 2010 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2010., 5551603, pp. 452-456, 2010 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2010, QingDao, China, 7/15/10. https://doi.org/10.1109/SOLI.2010.5551603
Peng J, Zeng D. Tag-based smoothing for item recommendation. In Proceedings of 2010 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2010. 2010. p. 452-456. 5551603 https://doi.org/10.1109/SOLI.2010.5551603
Peng, Jing ; Zeng, Dajun. / Tag-based smoothing for item recommendation. Proceedings of 2010 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2010. 2010. pp. 452-456
@inproceedings{5ba3d3e5934441c98fa3b1499063e2c5,
title = "Tag-based smoothing for item recommendation",
abstract = "Collaborative tagging sites allow users to save and annotate their favorite Web contents with tags. These tags provide a novel source of information for collaborative filtering. However, major gap exists in how to integrate tagging information into traditional recommender systems for better recommendation quality, due to the difficulty to quantize these semantic tags. This paper proposes a novel approach to convert this semantic information into quantitative values from a smoothing point of view, taking advantage of the topicbased method, and then make recommendation in a traditional user-based CF fashion based on the smoothed user-item matrix. Experiments on two real-world collaborative tagging datasets prove the effectiveness of our approach.",
keywords = "Collaborative filtering, Recommender systems, Smoothing, Tag, Weighing",
author = "Jing Peng and Dajun Zeng",
year = "2010",
doi = "10.1109/SOLI.2010.5551603",
language = "English (US)",
isbn = "9781424471188",
pages = "452--456",
booktitle = "Proceedings of 2010 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2010",

}

TY - GEN

T1 - Tag-based smoothing for item recommendation

AU - Peng, Jing

AU - Zeng, Dajun

PY - 2010

Y1 - 2010

N2 - Collaborative tagging sites allow users to save and annotate their favorite Web contents with tags. These tags provide a novel source of information for collaborative filtering. However, major gap exists in how to integrate tagging information into traditional recommender systems for better recommendation quality, due to the difficulty to quantize these semantic tags. This paper proposes a novel approach to convert this semantic information into quantitative values from a smoothing point of view, taking advantage of the topicbased method, and then make recommendation in a traditional user-based CF fashion based on the smoothed user-item matrix. Experiments on two real-world collaborative tagging datasets prove the effectiveness of our approach.

AB - Collaborative tagging sites allow users to save and annotate their favorite Web contents with tags. These tags provide a novel source of information for collaborative filtering. However, major gap exists in how to integrate tagging information into traditional recommender systems for better recommendation quality, due to the difficulty to quantize these semantic tags. This paper proposes a novel approach to convert this semantic information into quantitative values from a smoothing point of view, taking advantage of the topicbased method, and then make recommendation in a traditional user-based CF fashion based on the smoothed user-item matrix. Experiments on two real-world collaborative tagging datasets prove the effectiveness of our approach.

KW - Collaborative filtering

KW - Recommender systems

KW - Smoothing

KW - Tag

KW - Weighing

UR - http://www.scopus.com/inward/record.url?scp=77957804296&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77957804296&partnerID=8YFLogxK

U2 - 10.1109/SOLI.2010.5551603

DO - 10.1109/SOLI.2010.5551603

M3 - Conference contribution

AN - SCOPUS:77957804296

SN - 9781424471188

SP - 452

EP - 456

BT - Proceedings of 2010 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2010

ER -