Estratégia Híbrida para recomendação personalizada utilizando o Guia de Programação Eletrônico. / Hybrid Strategy for personalized recommendation using the Electronic Program Guide.
AUTOR(ES)
Mário Alves de Moraes Neto
FONTE
IBICT - Instituto Brasileiro de Informação em Ciência e Tecnologia
DATA DE PUBLICAÇÃO
31/03/2011
RESUMO
The wide range of digital information has become more difficult to find things that best fit the profile of those who demand something. In this context, Recommender Systems represent a solution to the above problem, since they help the user to find relevant information, operating based on information personalization, since it helps certain individuals in the social process of appointing and receiving an indication of items belonging to a specific domain. Similarly, the digitization of TV signals allows a wide range of channels can be transmitted, resulting the users face a range of television programming so extensive. While this represents a benefit, hence draw a problem: users with an expressive range of programs without knowing what those best fit on your respective profile. Therefore, making use of the benefits of Recommender Systems, this paper proposes the use of techniques for Content-Based Filtering (CBF) and the Association Rules to recommend programs in the context of interactive digital television. The system presented here, which allows create multiple profiles and use the format of an Electronic Program Guide (EPG), also allows the user to evaluate the recommendation process, making possible the continous improvement of this recommendation system.
ASSUNTO(S)
epg televisão digital ciencia da computacao regras de associação estratégia híbrida sistemas de recomendação recommender systems hybrid strategy epg digital tv association rules
