Analysis of online shopping reviews based on natural language processing – Taking Amazon as an example

Based on the idea of natural language processing, a Naive Bayesian Model is established, and the scoring formula based on star rating and user evaluation is determined, so as to calculate the scores of all users. It is proved that the product has a life cycle, and the product popularity will decrease with the passage of time. The value model of products is established, and the products are clustered to get the value score of each product, so as to judge the success or failure of a product. Use python to count the frequency of each star's comments. We choose specific words and analyze the internal relationship between them and star rating. Finally, the strategies of consumer marketing refining, opening “technology +” mode and putting forward upgrading marketing effectiveness in the whole cycle.