Conversion Prediction of a Bank Marketing Campaign

• Goal of the Project: To predict If a customer would subscribe marketing campaign of a Portugese Bank(Term Deposit) by building a supervised learning model(Random Forest with Hyper Parameter Tuning) and by this way reducing the costs.

• Dataset: UCI ML Repository. Data consists of 20 input features which are both numerical and categorical and related to customers and marketing campaign. Raw data has ~ 41k records. Marketing campaigns were based on phone calls, target variable represents if a customer subscribes a term deposit or not.

• High-Level Planing:

  1. Data Exploration
  2. Data Processing and Feature Selection
  3. Visualization and EDA
  4. Model Development
  5. Model Evaluation
  6. Model Deployment

• Results:

  89% Accuracy and 88% Recall Score on Test Set.