부산대학교 도서관

Predicting the purchase intention of customers is very important to the relevant decision-making departments of brands. This article aims to provide decision-making basis and suggestions for decision-making departments by analyzing the personal situation and related satisfaction of electric vehicle target customers. This article first classifies users based on systematic clustering, uses Spearman correlation coefficients, etc. to find factors that are strongly related to customers' purchase intentions. Secondly, this article uses logistic regression analysis and establishes a Bayesian discriminant model to predict the correctness of purchases. The results were compared, and the most accurate model was adopted to understand the customer's purchase intention; finally, the objective function was established and solved when the satisfaction degree was increased by 5%, This paper provides suggestions for car companies in terms of market positioning and service intensity.