Insurance Renewal Probability Model Based on Deep Learning Algorithm

Combined with the realistic background of vehicle insurance in daily life, this paper establishes a set of insurance renewal probability calculation model based on deep learning algorithm, and USES deep learning algorithm to analyze the data of users to calculate the renewal probability of users, which provides an effective reference method for the calculation of vehicle insurance renewal probability. This article USES Python and SPSS to analyze the collected data. Then the Isolation forest algorithm was used to analyze outliers. After extracting the outliers, Pearson correlation coefficient was used to analyze the correlation between the features of the data set and the renewal of the vehicle. The logistic regression model is established, the accurate portrait of the customer is made, and the probability of the customer's renewal is solved. According to the main impact of its renewal, the corresponding preferential policies have been formulated. It provides a certain basis for the future auto insurance industry to improve the customer renewal rate.