Emotional analysis model of college students based on big data public opinion
Download as PDF
DOI: 10.23977/ICSSEM2023.044
Corresponding Author
Shichao Luo
ABSTRACT
With the development of Chinese social media, the occurrence of major events and social disputes can quickly produce a large number of text data, and academic and industrial circles pay more and more attention to emotional analysis of public opinion events. The contents involved in university online public opinion are often the real-time reflection of students' various ideological trends. Because the psychology and physiology of college students are not mature enough, in the process of online public opinion events, extreme speech and behavior are often produced due to the stimulation of various external factors, which has a negative impact on the harmony and stability of universities. In this paper, an emotional analysis model of college students based on big data public opinion is designed. Through the technology of web crawler, Word2vec model is adopted to optimize the emotional dictionary, CNN-LSTM (Convolutional Neural Network-Long short-term memory) carries out text emotion classification module, and the emotional analysis of user comments is carried out. According to the analysis of students' public opinion, the effective prediction of college students' excessive behavior is realized, which significantly improves the quality of students' mental health.
KEYWORDS
Big data; Students; Emotional analysis; Public opinion