TY - GEN
T1 - Learning distributed representations from reviews for collaborative filtering
AU - Almahairi, Amjad
AU - Kastner, Kyle
AU - Cho, Kyunghyun
AU - Courville, Aaron
N1 - Publisher Copyright:
© 2015 ACM.
PY - 2015/9/16
Y1 - 2015/9/16
N2 - Recent work has shown that collaborative filter-based recommender systems can be improved by incorporating side information, such as natural language reviews, as a way of regularizing the derived product representations. Motivated by the success of this approach, we introduce two different models of reviews and study their effect on collaborative filtering performance. While the previous state-of-the-art approach is based on a latent Dirichlet allocation (LDA) model of reviews, the models we explore are neural network based: a bag-of-words product-of-experts model and a recurrent neural network. We demonstrate that the increased exibility offered by the product-of-experts model allowed it to achieve state-of-the-art performance on the Amazon review dataset, outperforming the LDA-based approach. However, interestingly, the greater modeling power offered by the recurrent neural network appears to undermine the model's ability to act as a regularizer of the product representations.
AB - Recent work has shown that collaborative filter-based recommender systems can be improved by incorporating side information, such as natural language reviews, as a way of regularizing the derived product representations. Motivated by the success of this approach, we introduce two different models of reviews and study their effect on collaborative filtering performance. While the previous state-of-the-art approach is based on a latent Dirichlet allocation (LDA) model of reviews, the models we explore are neural network based: a bag-of-words product-of-experts model and a recurrent neural network. We demonstrate that the increased exibility offered by the product-of-experts model allowed it to achieve state-of-the-art performance on the Amazon review dataset, outperforming the LDA-based approach. However, interestingly, the greater modeling power offered by the recurrent neural network appears to undermine the model's ability to act as a regularizer of the product representations.
KW - Deep learning
KW - Neural networks
KW - Recommender systems
UR - http://www.scopus.com/inward/record.url?scp=84962816292&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84962816292&partnerID=8YFLogxK
U2 - 10.1145/2792838.2800192
DO - 10.1145/2792838.2800192
M3 - Conference contribution
AN - SCOPUS:84962816292
T3 - RecSys 2015 - Proceedings of the 9th ACM Conference on Recommender Systems
SP - 147
EP - 154
BT - RecSys 2015 - Proceedings of the 9th ACM Conference on Recommender Systems
PB - Association for Computing Machinery, Inc
T2 - 9th ACM Conference on Recommender Systems, RecSys 2015
Y2 - 16 September 2015 through 20 September 2015
ER -