Research on the Influencing Factors of Fan Reviews and Sentiment of We-media in Social Commerce
Fan reviews of We-media in social commerce reflect the fans’ sentiment. Researching the number of fan reviews and the influential factors of fan sentiments can help to understand the fan sentiment characteristics and attract their attention. We use crawler software to crawl Li Ziqi’s fan reviews on Today’s Headlines, and use SnowNLP sentiment analysis to quantify the sentiment attributes of the reviews into sentiment indexes. We construct three types of features: interactive behavior, sentiment attributes and video attributes. We use regression method to analyze the relationship between the number of reviews and sentimental attributes and other characteristics. The results show that the number of fan reviews is directly proportional to the number of broadcasts, the number of likes, and time span; the average sentiment index is directly proportional to the time span and positive reviews, and inversely proportional to the video duration. The results provide a basis for social media and bloggers to manage fan reviews effectively.