A Novel Embedding Model for Relation Prediction in Recommendation Systems

Yu ZHAO; Sheng GAO; Patrick GALLINARI; Jun GUO

doi:10.1587/transinf.2016EDP7421

A Novel Embedding Model for Relation Prediction in Recommendation Systems

Yu ZHAO, Sheng GAO, Patrick GALLINARI, Jun GUO

Full Text Views

0

Share
Cite this

Summary :

It inevitably comes out information overload problem with the increasing available data on e-commence websites. Most existing approaches have been proposed to recommend the users personal significant and interesting items on e-commence websites, by estimating unknown rating which the user may rate the unrated item, i.e., rating prediction. However, the existing approaches are unable to perform user prediction and item prediction, since they just treat the ratings as real numbers and learn nothing about the ratings' embeddings in the training process. In this paper, motivated by relation prediction in multi-relational graph, we propose a novel embedding model, namely RPEM, to solve the problem including the tasks of rating prediction, user prediction and item prediction simultaneously for recommendation systems, by learning the latent semantic representation of the users, items and ratings. In addition, we apply the proposed model to cross-domain recommendation, which is able to realize recommendation generation in multiple domains. Empirical comparison on several real datasets validates the effectiveness of the proposed model. The data is available at https://github.com/yuzhaour/da.

Publication: IEICE TRANSACTIONS on Information Vol.E100-D No.6 pp.1242-1250

Publication Date: 2017/06/01

Publicized: 2017/03/14

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2016EDP7421

Type of Manuscript: PAPER

Category: Artificial Intelligence, Data Mining

Authors

Yu ZHAO
  Beijing University of Posts and Telecommunications
Sheng GAO
  Beijing University of Posts and Telecommunications
Patrick GALLINARI
  University of Paris VI
Jun GUO
  Beijing University of Posts and Telecommunications

Keyword

recommendation system, latent factor model, relation prediction, cross-domain recommendation

Cite this

Copy

Yu ZHAO, Sheng GAO, Patrick GALLINARI, Jun GUO, "A Novel Embedding Model for Relation Prediction in Recommendation Systems" in IEICE TRANSACTIONS on Information, vol. E100-D, no. 6, pp. 1242-1250, June 2017, doi: 10.1587/transinf.2016EDP7421.
Abstract: It inevitably comes out information overload problem with the increasing available data on e-commence websites. Most existing approaches have been proposed to recommend the users personal significant and interesting items on e-commence websites, by estimating unknown rating which the user may rate the unrated item, i.e., rating prediction. However, the existing approaches are unable to perform user prediction and item prediction, since they just treat the ratings as real numbers and learn nothing about the ratings' embeddings in the training process. In this paper, motivated by relation prediction in multi-relational graph, we propose a novel embedding model, namely RPEM, to solve the problem including the tasks of rating prediction, user prediction and item prediction simultaneously for recommendation systems, by learning the latent semantic representation of the users, items and ratings. In addition, we apply the proposed model to cross-domain recommendation, which is able to realize recommendation generation in multiple domains. Empirical comparison on several real datasets validates the effectiveness of the proposed model. The data is available at https://github.com/yuzhaour/da.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.2016EDP7421/_p

Copy

@ARTICLE{e100-d_6_1242,
author={Yu ZHAO, Sheng GAO, Patrick GALLINARI, Jun GUO, },
journal={IEICE TRANSACTIONS on Information},
title={A Novel Embedding Model for Relation Prediction in Recommendation Systems},
year={2017},
volume={E100-D},
number={6},
pages={1242-1250},
abstract={It inevitably comes out information overload problem with the increasing available data on e-commence websites. Most existing approaches have been proposed to recommend the users personal significant and interesting items on e-commence websites, by estimating unknown rating which the user may rate the unrated item, i.e., rating prediction. However, the existing approaches are unable to perform user prediction and item prediction, since they just treat the ratings as real numbers and learn nothing about the ratings' embeddings in the training process. In this paper, motivated by relation prediction in multi-relational graph, we propose a novel embedding model, namely RPEM, to solve the problem including the tasks of rating prediction, user prediction and item prediction simultaneously for recommendation systems, by learning the latent semantic representation of the users, items and ratings. In addition, we apply the proposed model to cross-domain recommendation, which is able to realize recommendation generation in multiple domains. Empirical comparison on several real datasets validates the effectiveness of the proposed model. The data is available at https://github.com/yuzhaour/da.},
keywords={},
doi={10.1587/transinf.2016EDP7421},
ISSN={1745-1361},
month={June},}

Copy

TY - JOUR
TI - A Novel Embedding Model for Relation Prediction in Recommendation Systems
T2 - IEICE TRANSACTIONS on Information
SP - 1242
EP - 1250
AU - Yu ZHAO
AU - Sheng GAO
AU - Patrick GALLINARI
AU - Jun GUO
PY - 2017
DO - 10.1587/transinf.2016EDP7421
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E100-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2017
AB - It inevitably comes out information overload problem with the increasing available data on e-commence websites. Most existing approaches have been proposed to recommend the users personal significant and interesting items on e-commence websites, by estimating unknown rating which the user may rate the unrated item, i.e., rating prediction. However, the existing approaches are unable to perform user prediction and item prediction, since they just treat the ratings as real numbers and learn nothing about the ratings' embeddings in the training process. In this paper, motivated by relation prediction in multi-relational graph, we propose a novel embedding model, namely RPEM, to solve the problem including the tasks of rating prediction, user prediction and item prediction simultaneously for recommendation systems, by learning the latent semantic representation of the users, items and ratings. In addition, we apply the proposed model to cross-domain recommendation, which is able to realize recommendation generation in multiple domains. Empirical comparison on several real datasets validates the effectiveness of the proposed model. The data is available at https://github.com/yuzhaour/da.
ER -