TY - GEN
T1 - Latent factor models and aggregation operators for collaborative filtering in reciprocal recommender systems
AU - Neve, James
AU - Palomares, Ivan
N1 - Funding Information:
This research has been supported by the EPSRC Doctoral Training Programme (DTP). The authors would also like to thank Eureka Inc. for providing the dataset to test our research and the resources to train the models, and in particular Mr. Shintaro Kaneko (CTO at Eureka Inc.) and Mr. Yusuke Usui (AI Team Leader at Eureka Inc.) for their support.
Publisher Copyright:
© 2019 Association for Computing Machinery.
PY - 2019/9/10
Y1 - 2019/9/10
N2 - Online dating platforms help to connect people who might potentially be a good match for each other. They have exerted a signifcant societal impact over the last decade, such that about one third of new relationships in the US are now started online, for instance. Recommender Systems are widely utilized in online platforms that connect people to people in e.g. online dating and recruitment sites. These recommender approaches are fundamentally diferent from traditional user-item approaches (such as those operating on movie and shopping sites), in that they must consider the interests of both parties jointly. Latent factor models have been notably successful in the area of user-item recommendation, however they have not been investigated within user-to-user domains as of yet. In this study, we present a novel method for reciprocal recommendation using latent factor models. We also provide a frst analysis of the use of diferent preference aggregation strategies, thereby demonstrating that the aggregation function used to combine user preference scores has a signifcant impact on the outcome of the recommender system. Our evaluation results report signifcant improvements over previous nearest-neighbour and content-based methods for reciprocal recommendation, and show that the latent factor model can be used efectively on much larger datasets than previous state-of-the-art reciprocal recommender systems.
AB - Online dating platforms help to connect people who might potentially be a good match for each other. They have exerted a signifcant societal impact over the last decade, such that about one third of new relationships in the US are now started online, for instance. Recommender Systems are widely utilized in online platforms that connect people to people in e.g. online dating and recruitment sites. These recommender approaches are fundamentally diferent from traditional user-item approaches (such as those operating on movie and shopping sites), in that they must consider the interests of both parties jointly. Latent factor models have been notably successful in the area of user-item recommendation, however they have not been investigated within user-to-user domains as of yet. In this study, we present a novel method for reciprocal recommendation using latent factor models. We also provide a frst analysis of the use of diferent preference aggregation strategies, thereby demonstrating that the aggregation function used to combine user preference scores has a signifcant impact on the outcome of the recommender system. Our evaluation results report signifcant improvements over previous nearest-neighbour and content-based methods for reciprocal recommendation, and show that the latent factor model can be used efectively on much larger datasets than previous state-of-the-art reciprocal recommender systems.
UR - http://www.scopus.com/inward/record.url?scp=85073358906&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85073358906&partnerID=8YFLogxK
U2 - 10.1145/3298689.3347026
DO - 10.1145/3298689.3347026
M3 - Conference contribution
AN - SCOPUS:85073358906
T3 - RecSys 2019 - 13th ACM Conference on Recommender Systems
SP - 219
EP - 227
BT - RecSys 2019 - 13th ACM Conference on Recommender Systems
PB - Association for Computing Machinery, Inc
T2 - 13th ACM Conference on Recommender Systems, RecSys 2019
Y2 - 16 September 2019 through 20 September 2019
ER -