Application of rough sets in audience rating prediction
Meimei Wu1, Yan Wang1, Xingli Liu2
COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(12C) 521-525
1School of Science, Communication University of China, Beijing 100024, China
2College of Information Science and Technology, Heilongjiang University of Science and Technology, Harbin, 150022, Heilongjiang, China
The Audience Rating Prediction plays a significant role in the increasingly fierce competition in the television industry. This paper proposes an approach of combining the rough sets with the back propagation neural network, which can be used to predict complicated audience ratings with dynamic and non-linear factors. The attribute reduction based on rough sets can remove redundant information, weaken the impacts of noise data and interdependency data to BP neural network and reduce the complexity of the neural network system. Therefore, this approach can improve the accuracy of prediction and reduce the training time. Through the experiments of audience rating data, this paper compares the approach based on Rough sets and BP neural network with that of BP neural network only. These experiments represent that the Audience rating prediction based on Rough sets and BP neural network achieves better results.