Hybrid recommendation for sparse rating matrix: A heterogeneous information network approach

Haiyang Zhang; Ivan Ganchev; Nikola S. Nikolov; Zhanlin Ji; Mairtin O'Droma

doi:10.1109/IAEAC.2017.8054114

Hybrid recommendation for sparse rating matrix: A heterogeneous information network approach

Haiyang Zhang, Ivan Ganchev, Nikola S. Nikolov, Zhanlin Ji, Mairtin O'Droma

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

10 Citations (Scopus)

Abstract

Exploiting additional item meta-data is proposed in this paper for solving data sparsity and cold start problems found in item-based collaborative filtering (CF) techniques, which are employed in recommendation systems. Additional item meta-data provides the foundation for generating a heterogeneous information network (HIN). The proposed approach is to enrich the item-based CF with diverse types of relationships existing between items in the HIN, to overcome the sparsity issue from implicit user feedback. Bayesian personalized ranking optimization technique is used for estimation and its performance is evaluated by comparing the results with the traditional item-based CF. The experimental tests prove that the proposed approach achieves better accuracy.

Original language	English
Title of host publication	Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017
Editors	Bing Xu
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	740-744
Number of pages	5
ISBN (Electronic)	9781467389778
DOIs	https://doi.org/10.1109/IAEAC.2017.8054114
Publication status	Published - 29 Sept 2017
Externally published	Yes
Event	2nd IEEE Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017 - Chongqing, China Duration: 25 Mar 2017 → 26 Mar 2017

Publication series

Name	Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017

Conference

Conference	2nd IEEE Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017
Country/Territory	China
City	Chongqing
Period	25/03/17 → 26/03/17

Keywords

Bayesian personalized ranking
Collaborative filtering
Heterogeneous information network
Meta-path
Recommendation system

Access to Document

10.1109/IAEAC.2017.8054114

Cite this

Zhang, H., Ganchev, I., Nikolov, N. S., Ji, Z., & O'Droma, M. (2017). Hybrid recommendation for sparse rating matrix: A heterogeneous information network approach. In B. Xu (Ed.), Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017 (pp. 740-744). Article 8054114 (Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IAEAC.2017.8054114

Zhang, Haiyang ; Ganchev, Ivan ; Nikolov, Nikola S. et al. / Hybrid recommendation for sparse rating matrix : A heterogeneous information network approach. Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017. editor / Bing Xu. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 740-744 (Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017).

@inproceedings{aa166d60072949a9b72b1e8477662258,

title = "Hybrid recommendation for sparse rating matrix: A heterogeneous information network approach",

abstract = "Exploiting additional item meta-data is proposed in this paper for solving data sparsity and cold start problems found in item-based collaborative filtering (CF) techniques, which are employed in recommendation systems. Additional item meta-data provides the foundation for generating a heterogeneous information network (HIN). The proposed approach is to enrich the item-based CF with diverse types of relationships existing between items in the HIN, to overcome the sparsity issue from implicit user feedback. Bayesian personalized ranking optimization technique is used for estimation and its performance is evaluated by comparing the results with the traditional item-based CF. The experimental tests prove that the proposed approach achieves better accuracy.",

keywords = "Bayesian personalized ranking, Collaborative filtering, Heterogeneous information network, Meta-path, Recommendation system",

author = "Haiyang Zhang and Ivan Ganchev and Nikolov, {Nikola S.} and Zhanlin Ji and Mairtin O'Droma",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 2nd IEEE Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017 ; Conference date: 25-03-2017 Through 26-03-2017",

year = "2017",

month = sep,

day = "29",

doi = "10.1109/IAEAC.2017.8054114",

language = "English",

series = "Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "740--744",

editor = "Bing Xu",

booktitle = "Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017",

}

Zhang, H, Ganchev, I, Nikolov, NS, Ji, Z & O'Droma, M 2017, Hybrid recommendation for sparse rating matrix: A heterogeneous information network approach. in B Xu (ed.), Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017., 8054114, Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017, Institute of Electrical and Electronics Engineers Inc., pp. 740-744, 2nd IEEE Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017, Chongqing, China, 25/03/17. https://doi.org/10.1109/IAEAC.2017.8054114

Hybrid recommendation for sparse rating matrix: A heterogeneous information network approach. / Zhang, Haiyang; Ganchev, Ivan; Nikolov, Nikola S. et al.
Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017. ed. / Bing Xu. Institute of Electrical and Electronics Engineers Inc., 2017. p. 740-744 8054114 (Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017).

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

TY - GEN

T1 - Hybrid recommendation for sparse rating matrix

T2 - 2nd IEEE Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017

AU - Zhang, Haiyang

AU - Ganchev, Ivan

AU - Nikolov, Nikola S.

AU - Ji, Zhanlin

AU - O'Droma, Mairtin

PY - 2017/9/29

Y1 - 2017/9/29

N2 - Exploiting additional item meta-data is proposed in this paper for solving data sparsity and cold start problems found in item-based collaborative filtering (CF) techniques, which are employed in recommendation systems. Additional item meta-data provides the foundation for generating a heterogeneous information network (HIN). The proposed approach is to enrich the item-based CF with diverse types of relationships existing between items in the HIN, to overcome the sparsity issue from implicit user feedback. Bayesian personalized ranking optimization technique is used for estimation and its performance is evaluated by comparing the results with the traditional item-based CF. The experimental tests prove that the proposed approach achieves better accuracy.

AB - Exploiting additional item meta-data is proposed in this paper for solving data sparsity and cold start problems found in item-based collaborative filtering (CF) techniques, which are employed in recommendation systems. Additional item meta-data provides the foundation for generating a heterogeneous information network (HIN). The proposed approach is to enrich the item-based CF with diverse types of relationships existing between items in the HIN, to overcome the sparsity issue from implicit user feedback. Bayesian personalized ranking optimization technique is used for estimation and its performance is evaluated by comparing the results with the traditional item-based CF. The experimental tests prove that the proposed approach achieves better accuracy.

KW - Bayesian personalized ranking

KW - Collaborative filtering

KW - Heterogeneous information network

KW - Meta-path

KW - Recommendation system

UR - http://www.scopus.com/inward/record.url?scp=85034602455&partnerID=8YFLogxK

U2 - 10.1109/IAEAC.2017.8054114

DO - 10.1109/IAEAC.2017.8054114

M3 - Conference Proceeding

AN - SCOPUS:85034602455

T3 - Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017

SP - 740

EP - 744

BT - Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017

A2 - Xu, Bing

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 25 March 2017 through 26 March 2017

ER -

Zhang H, Ganchev I, Nikolov NS, Ji Z, O'Droma M. Hybrid recommendation for sparse rating matrix: A heterogeneous information network approach. In Xu B, editor, Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 740-744. 8054114. (Proceedings of 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2017). doi: 10.1109/IAEAC.2017.8054114

Hybrid recommendation for sparse rating matrix: A heterogeneous information network approach

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this