Design and Implementation of Sentiment Analysis of Weibo Comments Based on Deep Learning

In order to solve the problem of the outbreak of network public opinions during the use of Weibo, this paper designs and implements a sentiment analysis model of Weibo comments text based on Bi-LSTM (Bidirectional LSTM, Bi_LSTM) and attention mechanism. The Bi-LSTM model learns contextual feature and the attention mechanism learns key feature. The experimental results show that the Precision, Recall and F1 of the model designed in this paper are 78.26%, 78.24% and 78.24% on the public data set weibo_senti_100k. It can effectively analyze the emotional tendencies of Weibo comment text.