Time series predict customer behavior analysis mobile banking

The purpose of this paper is to help businesses use data mining to analyze customers dynamic behavior in mobile banking and then drive augmented customer relationships over time. The tool is used for customer relationships Traditionally, customer segmentation strategies are employed when dealing with a large community of customers. The research recommends an innovative methodology for forecasting segment-level customer behavior. This method makes use of both the equipment and methods used by forecasters, such as computational intelligence methods and traditional time series forecasting methods. An assessment of the suitability of the suggested method is conducted, as well as a case study based on data from bank customers mobile banking. The result of the ARMA method in the RMSE evaluation model is 0.37 and the confusion matrix has an accuracy of 0.89 for KNN (K = 1, lag-1:4).