A Method of Emotional Analysis of Movie Based on Convolution Neural Network and Bi-directional LSTM RNN

Abstract

The movie recommendation system provides an important way for users to choose a movie. In order to avoid manmode operations in the movie recommendation, the emotional analysis in the recommended system provides a new way of thinking. The traditional methods were based on knowledge, statistics and the hybrid approach. However these methods could obtain effect only base on the dataset with small amount and the one with not rich semantics. Therefore, in this paper we propose a new method based on convolution neural network and bi-directional LSTM RNN for dataset with large amount and rich semantics. Our approach does not need artificial labeled data and syntactic analysis, and only uses a small labeled train dataset. Also we prove the effectiveness and the accuracy of the proposed method by comparing with the current models including CNN, LSTM and Bi-LSTM. The experimental results demonstrate the effectiveness of the proposed method.