Using Visual Features and Latent Factors for Movie Recommendation

Published in Workshop on New Trends in Content-based Recommender Systems as part of the 10th ACM Conference of Recommender Systems, 2016

Recommended citation: Yashar Deldjoo, Mehdi Elahi, Paolo Cremonesi Workshop on New Trends in Content-based Recommender Systems, 2016 (CBRecSys@RecSys 2016).

[Proccedings] [PDF] [bibtex]


Item features play an important role in movie recommender systems, where recommendations can be generated by using explicit or implicit preferences of users on attributes such as genres. Traditionally, movie features are human-generated, either editorially or by leveraging the wisdom of the crowd. In this short paper, we present a recommender system for movies based of Factorization Machines that makes use of the low-level visual features extracted automatically from movies as side information. Low-level visual features – such as lighting, colors and motion – represent the design aspects of a movie and characterize its aesthetic and style. Our experiments on a dataset of more than 13K movies show that recommendations based on low-level visual fea- tures provides almost 10 times better accuracy in compar- ison to genre based recommendations, in terms of various evaluation metrics.