The Relationship Between Social Media Comments And New Energy Car Sales

In recent years,as the concept of "green development" has deepened into all aspects of China’s economic development,the new energy passenger car industry has developed at a rapid pace in the global context of energy conservation and emission reduction.At the same time,the Internet as an information dissemination medium has the advantage of fast and timely information dissemination in the field of public information,and social media users can express their inclination and attitude towards certain social hot issues.Therefore,reasonable monitoring and analysis of online public opinion in the new energy passenger car industry is regarded as an important direction for the future development of the passenger car industry.In this thesis,we use text mining techniques to identify the concerns of new energy vehicle users about passenger cars and explore the variable relationships between semantic resonance,emotional polarity,group heterogeneity and new energy passenger car sales in user opinion.The thesis firstly crawled the user comments of the new energy passenger car section from April 2020 to March 2022 on mainstream car review websites based on web crawlers and text mining techniques and used jieba word splitting technology,TF-IDF word frequency statistics and LDA topic model to analyse the user comments.The results show that users are generally satisfied with the new energy passenger cars,but there is still room for improvement in terms of sound insulation and interior trim.Secondly,this thesis calculates the measures of semantic resonance,emotional polarity and group heterogeneity in users’ opinions based on RelCat Tree,the text sentiment analysis of BosonNLP sentiment dictionary and the first-order and second-order entropy theories.The differences in the direction,degree and duration of influence of the three user opinion variables on the sales of new energy passenger cars were analysed by establishing vector error correction(VECM)models,as well as impulse response functions and variance decomposition.The main findings of the empirical study are:(ⅰ)Semantic resonance is an important factor influencing new energy passenger car sales,and this influence is negative.(ⅱ)Affective polarity has an impact on sales to some extent,and there is a negative interaction between the two.(ⅲ)Group heterogeneity as a measure of the non-homogeneity of network nodes indirectly affects user opinion and thus has an impact on sales.Finally,based on the results of the above model analysis,feasible suggestions for the new energy passenger car industry are put forward,including infrastructure construction,strengthening the detection of user opinion,and strengthening the ability to disseminate benign public opinion information.