A recommendation approach for digital TV systems based on multimodal features

Digital TV content providers are becoming widespread, with hundreds of programs available each day. The information overload makes difficult for the user to find programs of interest. To help the user, Recommendation Systems (RS) are a popular path. However, applying RS to some environments is not easy, either due to the lack or insufficiency of data to create accurate recommendations. In Digital TV domain, the main information available to make recommendations is the Electronic Program Guide (EPG) that is limited, containing only reduced textual data, making difficult to get an accurate recommendation using standard techniques. In this work we introduce a multimodal approach to recommend Digital TV programs, combining EPG text and visual information. We experimentally demonstrated that using multimodal features improved accuracy when compared with RS standard approaches.