Some details
Goal: to develop prediction algorithms:
1. Total customer spending for the next 30 days for each service/product category
2. Individual customer spending in the next month for each service/product category
Source data: 7 mln card transactions by 15K customers, 184 categories of services/products
Result:
• For Goal 1, an ensemble of models was a solution: time series prediction, linearregression, decision trees
• For Goal 2, we built a model based on statistics for prior periods and extended statistics on various combination of attributes
