Sentiment Analysis on Power Rationing Micro Blog Comments Based on SnowNLP-SVM-LDA Model

Abstract

In the context of energy conservation and emission reduction, many places across the country have introduced "power rationing" policies. In order to understand the public's emotional attitude and emotional characteristics of the policy, help government departments to grasp the trend of public opinion and optimize strategic decision-making, this paper proposes a public opinion analysis method based on the SnowNLP-SVM-LDA model. This paper takes micro blog comments on power rationing as the research object, and obtains a total of 5242 micro blog comments through Python collection and data cleaning. Firstly, the word cloud map is drawn for the comment text after Jieba word segmentation, and then the SnowNLP-SVM model is used to classify the sentiment of micro blog comments, and then LDA topic mining is carried out for positive and negative comments respectively.Research shows that most micro blog netizens (69.34% of the total sample) hold a negative attitude towards power cuts. This paper identifies three major topics of national policy, support for the government, and power supply security in positive emotions, and three major topics of people's livelihood issues, rising coal prices, and government work in negative emotions. Through the analysis of this paper, it provides data support and theoretical support for government departments to guide the trend of public opinion and optimize work decisions.