- Analyze user data to identify users with higher intention to purchase classes and forward them totelemarketing department. - Collect raw data for data analysis and processing from log files using MySQL. - Analyze the correlation between features and user pruchase behavior and explore new feature ofexpression from raw data files using Python and statiscal analysis methods. - Build the corresponding algorithm models, including XGBoost, Wide & Deep, etc. optimize the modelstructure and paramters to increase AUC. - Analyze user purchase rate based on backstage data reports. select bad cases and analyze thereasons for negative campaigning effect