Affiliations: School of Accounting & Finance, Xi’an Peihua University, Xi’an, China
Correspondence:
[*]
Corresponding author. Xiaoyi Yuan, School of Accounting & Finance, Xi’an Peihua University, Xi’an 710125, China. E-mail:1161232781@qq.com.
Abstract: The lack of effective evaluation of online education is a worldwide malpractice, and it is impossible to help students improve the correctness of online learning choices through existing reviews. Based on the current mainstream sentiment lexicon and text sentiment analysis, the authors use machine learning method to analyze the sentiment orientation of the legal course review text, through method that combines PMI and SVM. At the same time, this paper uses LibSVM tool to train and predict data, collect and pre-process data through network data collection, and, based on traditional algorithms, propose improved experimental scheme based on their respective advantages and disadvantages. In addition, the model proposed in this study is used to classify and process the emotional text, and the two methods are combined to obtain the final result. Finally, this paper combines experiments to analyze the performance of the comprehensive model proposed in this study. The research shows that the classification effect of the text sentiment analysis of model is good, it can be applied to practice, and it can provide theoretical reference for subsequent related research.
Keywords: Support vector machine, algorithmic optimization, online course, data
network
