Unsupervised TV Scheduling Based on Collaborative Filtering

Abstract

The generation of a television program schedule is a daily challenge. Broadcasting companies need to broadcast the best content according to the audience's preference. This paper proposes a novel take on collaborative filtering techniques to automate the selection of content based on viewers' rating through an interactive digital TV application. In order to overcome the fact that a television channel can broadcast only one schedule at a time, we group all ratings on Time Intervals in order to generalize viewers' preference and present the most appropriate content using a collaborative filtering technique.