Personalized TV program guide based on neural network

Author

Krstic, Miroslav ; Bjelica, M.

Author_Institution

Sch. of Electr. Eng. (ETF), Commun. Dept., Univ. of Belgrade, Belgrade, Serbia

fYear

2012

fDate

20-22 Sept. 2012

Firstpage

227

Lastpage

230

Abstract

As digital TV providers today offer hundreds of channels, TV viewers do not have problem with content availability, but with finding an interesting content in a reasonable time instead. In a situation like this, both the providers and the viewers would benefit from personalized TV program guides, the tools that would track and learn the viewers´ preferences and then recommend them the content they would like. In this paper, we propose one such guide which is based on artificial neural network. We examine and compare several learning algorithms, with recommendation accuracy and neural network training time as performance metrics.

Keywords

digital television; learning (artificial intelligence); neural nets; recommender systems; artificial neural network training time; content availability; digital TV providers; digital TV viewers; learning algorithms; performance metrics; personalized TV program guide; viewer preference learning; viewer preference recommendation accuracy; viewer preference tracking; Accuracy; Classification algorithms; Motion pictures; Neural networks; Recommender systems; TV; Training; Digital TV; feedforward neural networks; personalized program guide; recommender systems; supervised learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Network Applications in Electrical Engineering (NEUREL), 2012 11th Symposium on

Conference_Location

Belgrade

Print_ISBN

978-1-4673-1569-2

Type

conf

DOI

10.1109/NEUREL.2012.6420017

Filename

6420017