Transfer learning for collaborative filtering via a rating-matrix generative model

Li, Bin; Yang, Qiang; Xue, Xiangyang

doi:10.1145/1553374.1553454

Cited by 294 publications

(176 citation statements)

References 9 publications

Supporting

Mentioning

175

Contrasting

Order By: Relevance

“…Some models have been proposed to provide jointly diverse recommendations of items belonging to multiple domains [17]. According to the strategies of exploiting knowledge, the crossdomain recommendation approaches can be classified into two categories: 1) aggregating knowledge [24] and 2) linking and transferring knowledge [9], [27].…”

Section: Related Workmentioning

confidence: 99%

A Novel Embedding Model for Relation Prediction in Recommendation Systems

Gao

Gallinari

Guo

2017

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYIt 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.

show abstract

Section: Related Workmentioning

confidence: 99%

A Novel Embedding Model for Relation Prediction in Recommendation Systems

Gao

Gallinari

Guo

2017

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…The survey was done on [7], [8] and [9] to understand the methods of transferring the knowledge across the domain. These Transfer learning methods are used to first individually collect the ratings matrix for each auxiliary input domain and then transferring the collective ratings to the target domain.…”

Section: Literature Reviewmentioning

confidence: 99%

Enhanced Cross Domain Recommender System using Contextual parameters in Temporal Domain

Joshi¹,

Patil²

2017

IJCATR

View full text Add to dashboard Cite

Cross-domain collaborative filtering (CDCF) is an evolving research topic in the modern recommender systems. Its main objective is to alleviate the data sparsity problem in individual domains by mixing and transferring the knowledge among the related domains. But there is also an issue of user interest drift over time because user's taste keeps on changing over time. We should consider various temporal domains to overcome user interest drift over time problem to predict more accurately as per the user's current interest. This paper discusses how to achieve effective ratings recommendations using contextual parameters in temporal domain in this research line. It calculates the contextual parameters as per user's current timestamp. This will enhance the recommendations more in line with the current temporal domain. It also deals with cross domain recommendations for both movies and novels based on their categories and similarities.

show abstract

“…However, TCF neglects the shared knowledge in rating patterns. Recently cross-domain recommendation methods also work on transferring knowledge of similar rating patterns, which have demonstrated that related domains do share a certain part of rating patterns [41,127,44,42,43,45]. For example, Gao et al [44] proposed a cluster-level latent feature model to learn the shared part of rating patterns of user groups.…”

Section: Background and Motivationmentioning

confidence: 99%

“…Among them, Li et al used a latent feature space to capture the comment rating patterns of user ratings [41,127]. However, Geo et al pointed out that in practice, sometimes there may not always be a common rating pattern across the related domains [44].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Regularized deep learning for fashion and social media data analysis

Jiang¹

View full text Add to dashboard Cite

Transfer learning for collaborative filtering via a rating-matrix generative model

Cited by 294 publications

References 9 publications

A Novel Embedding Model for Relation Prediction in Recommendation Systems

A Novel Embedding Model for Relation Prediction in Recommendation Systems

Enhanced Cross Domain Recommender System using Contextual parameters in Temporal Domain

Regularized deep learning for fashion and social media data analysis

Contact Info

Product

Resources

About