Sentiment analysis of commodity reviews based on multilayer LSTM network
Hao Yan Sun, Tao Jiang, Yugang Dai · 2019
With the rapid development of network technology and the popularization and application of e-commerce, comments with subjective sentiments on various mainstream e-commerce platforms play a crucial role in improving product quality. This paper takes the commodity comments on e-commerce websites as the research object, studies the sentimental tendency of the comment text, expresses the comment text based on the word2vec word vector model, and establishes the sentimental analysis model of the deep hierarchical network structure by adopting the multi-layer LSTM. Experiments show that the sentiment analysis method proposed in this paper is obviously superior to the traditional machine learning method and convolutional neural network method, and the accuracy rate reaches 92.23%, and model have a small time cost. This method can be applied to the comment sentiment analysis in the field of e-commerce and network social platforms.