Electric Vehicle Sales Analysis Model Based on User Purchase Intention Analysis

Abstract

New energy vehicles are a major innovation field in China's 12th Five-Year Plan.In order to study the sales strategy of target customers of electric vehicles, this paper establishes the sales prediction model of electric vehicles. First, the initial decision tree model of customer mining of different brands of electric vehicles is constructed.The model was optimized by changing the minimum sample number of leaf nodes and pruning.At the same time, we do cross validation error analysis on three decision tree models, and finally get a stable and sensitive customer mining model.In the evaluation of the model, four indicators TP, FN, FP and TN in the confusion matrix were used to calculate the classification accuracy, recall rate, precision rate and F1 score, and the calculation results showed that the model had good classification and prediction effect.