We describe a novel, multi-task recommendation model, which jointly learns to perform rating prediction and recommendation explanation by combining matrix factorization, for rating prediction, and adversarial sequence to sequence learning for explanation generation. The result is evaluated using real-world datasets to demonstrate improved rating prediction performance, compared to state-of-the-art alternatives, while producing effective, personalized explanations.
%0 Conference Paper
%1 Lu:2018:WIL:3240323.3240365
%A Lu, Yichao
%A Dong, Ruihai
%A Smyth, Barry
%B Proceedings of the 12th ACM Conference on Recommender Systems
%C New York, NY, USA
%D 2018
%I ACM
%K explanation recommender recsys2018
%P 4--12
%R 10.1145/3240323.3240365
%T Why I Like It: Multi-task Learning for Recommendation and Explanation
%U http://doi.acm.org/10.1145/3240323.3240365
%X We describe a novel, multi-task recommendation model, which jointly learns to perform rating prediction and recommendation explanation by combining matrix factorization, for rating prediction, and adversarial sequence to sequence learning for explanation generation. The result is evaluated using real-world datasets to demonstrate improved rating prediction performance, compared to state-of-the-art alternatives, while producing effective, personalized explanations.
%@ 978-1-4503-5901-6
@inproceedings{Lu:2018:WIL:3240323.3240365,
abstract = {We describe a novel, multi-task recommendation model, which jointly learns to perform rating prediction and recommendation explanation by combining matrix factorization, for rating prediction, and adversarial sequence to sequence learning for explanation generation. The result is evaluated using real-world datasets to demonstrate improved rating prediction performance, compared to state-of-the-art alternatives, while producing effective, personalized explanations.},
acmid = {3240365},
added-at = {2018-12-25T18:59:29.000+0100},
address = {New York, NY, USA},
author = {Lu, Yichao and Dong, Ruihai and Smyth, Barry},
biburl = {https://www.bibsonomy.org/bibtex/25fda5aa9eef1fbc8e383b0ebc42dda06/brusilovsky},
booktitle = {Proceedings of the 12th ACM Conference on Recommender Systems},
description = {Why I like it},
doi = {10.1145/3240323.3240365},
interhash = {bc2f2e9d1a98d8171de7e0006b5233eb},
intrahash = {5fda5aa9eef1fbc8e383b0ebc42dda06},
isbn = {978-1-4503-5901-6},
keywords = {explanation recommender recsys2018},
location = {Vancouver, British Columbia, Canada},
numpages = {9},
pages = {4--12},
publisher = {ACM},
series = {RecSys '18},
timestamp = {2019-06-09T09:32:34.000+0200},
title = {Why I Like It: Multi-task Learning for Recommendation and Explanation},
url = {http://doi.acm.org/10.1145/3240323.3240365},
year = 2018
}