Campaign Optimization - Predicting How and When to Contact Leads

SQL Database Tables


Below are the different data sets that you will find in your database after deployment.

Table Description
Campaign_Detail Raw data about each marketing campaign that occurred
Lead_Demography Raw demographics and financial data about each customer
Market_Touchdown Raw channel-day-time data used for every customer of Lead_Demography in every campaign he was targeted
Product Raw data about the product marketed in each campaign
CM_AD0 Merged data with missing values replaced by the mode
CM_AD Analytical data set: aggregated cleaned data set with new features
CM_AD_N Normalized version of CM_AD
All tables with the _N (e.g. Age_N) Table holding the normalization correspondance between the factors of each variable and the assigned integer
Train_Id Lead_Id that will go to the training set
CM_AD_Train Training set
CM_AD_Test Testing set
Campaign_Models Table storing the trained models
Forest_Prediction Prediction results for testing step with Random Forest
Boosted_Prediction Prediction results for testing step with GBT
Campaign_Metrics Performance metrics of the models tested
Best_Model Table with the name of the best model based on AUC
Prob_Id Table with the scoring results on the full table with the best model
Recommendations Table with the recommendations for each customer