##Description This is a demo of dynamic customer targeting. Treating the marketing customer targeting as stream process, we can find out potential buyes with less marketing expenses.
##Result With a simple Naive Bayes model, we can save 25% - 30% calls in a marketing campaign.
##Structure
- data_preparation.R for pre processing the raw data
- modeling_xxx.R for fitting the model
- Summary.R for model comparation
- Incremental_model.R a recursive method to do customer targeting
##Data Source: [Moro et al., 2014] S. Moro, P. Cortez and P. Rita. A Data-Driven Approach to Predict the Success of Bank Telemarketing. Decision Support Systems, Elsevier, 62:22-31, June 2014 https://archive.ics.uci.edu/ml/datasets/Bank+Marketing