Mining model of purchase intention based on support vector machine algorithm

The automobile industry is an important pillar industry of the national economy, while new energy automobile industry is a strategic emerging industry. Firstly, the Spearman correlation coefficient of each index of each brand is calculated, and then the influencing factors of each brand are obtained, and the customer mining model of each brand is established. Finally, it tests the willingness of 15 target customers to buy electric vehicles and conducts k-fold cross-verification. The results show that the purchase user's number of different brands is NO.2 and 4 for brand 1, NO.7 for brand 2 and NO.12 for brand 3. The comprehensive prediction accuracy is as high as 0.94, indicating that the model has high sensitivity and the test results are accurate.