In this practical application, we explored the data from a Portugese banking institution and is a collection of the results of multiple marketing campaigns. The dataset collected is related to 17 campaigns that occurred between May 2008 and November 2010, corresponding to a total of 79354 contacts.
In this analysis, I compare several classifiers including K-Nearest Neighbors, Logistic Regression, Decision Trees, and Support Vector Machines. The objective is to predict the outcome of a bank promotions campaign for individuals.
The business objective is to find a model that explains the success of a contact, improving campaign efficiency and resource management while selecting potential buying customers effectively.
We applied the CRISP-DM framework for understading the data, cleaning data and applying various models for the data.
We started with a basic dummy model and extended basic version of the models like K Nearest Neighbor, Logistic Regression, Decision Trees, and Support Vector Machines. Later, fine tuned them further with GridSearchAV hyper tuning paramters.
In conclusion, when compared with train time, train accuracy and test accuracy, we found that SVM (linear) model performed better than other models.