Framework for prediction of life insurance customers based on multi-algorithms

This paper's objective is to propose a framework for the prediction of life insurance customers based on multi-algorithms. The framework consists of three modules. The first is the data preparation module. The second is the data cleaning module. The third is the data extraction module. The process of data extraction is divided into 3 steps. The first step is feature selection. This step select the optimal feature for data analysis. The second step cluster the data by utilizing K-mean algorithm. The last step extract data based on a neural network algorithm in order to create a recommendation model. The comparison results show that the use of multi-algorithms attained the highest predictive accuracy at 92.83%.