Contextual TV ow commendation
Paula G´ omez DURAN
1
, Jordi VITRI
`
A
Departament de Matem ` atiques i Inform ` atica, Universitat de Barcelona
Abstract.
Recommender systems are a form of artificial intelligence that is used to sug-
gest items to users of digital platforms. They use large data sets to infer models of
users’ behavior and preferences in order to recommend items that the user may be
interested in. Following the trend imposed by digital media companies and willing
to adapt to the media consumption habits of their customers, TV broadcasters are
starting to realize the potential of recommender systems to personalize the access
to their online catalog. By understanding what viewers are watching and what they
might like, TV broadcasters can improve the quality of their programming, increase
viewership, and attract new viewers.
In this work, we analyze one specific group of users that TV broadcasters must
take into account when creating a recommender system: non-logged users. In this
scenario the challenge is to use contextual information about the interaction in order
to predict recommendations, as it is not feasible to use any kind of information
about the user. We propose a method to leverage data from other type of users
(logged users and identified devices) by using Graph Convolutional Networks in
order to come up with a more accurate recommender system for unidentified users.
Keywords. Recommender Systems, Graph Convolutional Network, Context-aware
recommendations, Public Media Service
1. Introduction
Recommender systems are a family of artificial intelligence tools that are used to suggest
items for users of digital platforms, such as online media (OM) platforms and public-
service media (PSM) platforms. In this scenario, they are specifically used to optimize
user’s engagement. Collaborative filtering has been identified as one of the best technical
approaches to accomplish this task [3, 5, 19], but the case of PSM platforms faces some
specific requirements, such as the presence of a special group of users that are not present
in OM and which we aim to analyse in this work: non-identified users. This kind of users
is very typical of PSMs, as on most of these platforms there is no requirement for a user
to be registered and cookies cannot always be guaranteed to track users’ sessions.
Over the last recent years, context-aware recommendations [11] have aroused in
order to end up with more accurate recommendations. Context-aware recommendations
[1] take into account all aspects surrounding a user’s situation when making suggestions.
1
Corresponding Author: Paula G´ omez Duran, Gran Via de les Corts Catalanes 585, 08007, Barcelona, Spain;
E-mail: paula.gomez@ub.edu.
Sh Re
Artificial Intelligence Research and Development
A. Cortés et al. (Eds.)
© 2022 The authors and IOS Press.
This article is published online with Open Access by IOS Press and distributed under the terms
of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).
doi:10.3233/FAIA220325
125