Sentiment Analysis of Weibo Platform Based on LDA-SnowNLP Model
Sicheng Lu, Qiang Liu, Zheyan Zhang
Abstract
In order to perform sentiment analysis on Weibo platform posts, a sentiment analysis model combining the LDA model and the machine learning model Snownlp is used to analyze posts data. Python is used to crawl posts with specific keywords within a specified time range. TF-IDF is used for high-frequency word statistics, the LDA model is employed for topic extraction, and the Snownlp model is used for sentiment analysis. Experimental results demonstrate that the model has a relatively clear restorative effect on the topic evolution in Weibo posts. Furthermore, the use of sentiment scores in a line graph intuitively displays the trend of sentiment attitudes in a specific time period.
Topics & Concepts
Sentiment analysisComputer sciencePython (programming language)Artificial intelligenceNatural language processingInformation retrievalOperating systemSentiment Analysis and Opinion MiningComputational and Text Analysis MethodsAdvanced Text Analysis Techniques