Bank Precision Marketing Solutions-- using Logistic Regression and Tree Algorithms Precision marketing makes sense for sellers and banks. Selecting a suitable user group for promotion, on the one hand, reduces the cost of promotion, and on the other hand increases the possibility of promotion success. In this project, we use a dataset from the Bank Precision Marketing Solutions competiton on kesci and predict the probability of user purchase.
The dataset used for this project is the online Dataset which contains property information of 25317 clients.
Logistic Regression; Support Vector Machine; Decision Trees; Random Forest;
To evaluate the performance of each model, we used the ROC AUC Score. Learn more about ROC curves and AUC here.
AUC of different models
