Sentiment Analysis of English Text based on LSTM-RNN

Abstract

<p style="text-align: justify;"><span style="font-size:16px;font-family: 'Times New Roman','serif'">With the emergence of a large number of text information on the Internet, not only can we learn the user's personal preferences from the emotional mining and analysis of the text, but also can monitor public opinion. So text sentiment analysis has become a new research hotspot for researchers on Web-based information processing. Aiming at the problem of how to analyze the sentiment class from the English text more accurately, this paper selects some user reviews dataset as the sentiment analysis database of the English text dataset by using the corpus that has already completed emotional annotation on the Internet as corpus, and uses the LSTM-RNN algorithm to classify and analyze. The analysis results of this method in this paper are compared with the traditional classification results of other models, the experimental results show that the LSTM-RNN model proposed in this paper compared with other traditional network models is more effective well in classifying sentiment analysis of English texts. The feature selection method of information gain can reach 88.2%, which proves that the model is effective.</span>