Predicting Term Deposit Subscriptions

Project Description

This project focuses on predicting whether bank clients will subscribe to term deposits using advanced Machine Learning techniques. The dataset used contains information from previous marketing campaigns, as well as socioeconomic data and client characteristics.

Project Structure

• 2410_kaggle_bank_campaingMarcelo.ipynb: Main notebook containing exploratory data analysis, feature engineering, model training, and evaluation.
• submissions/: Folder containing various predictions generated by the trained models.
• test.csv: Test dataset used to evaluate the models.
• train.csv: Training dataset used to train the models.

Notebook Content

1. Working Environment Setup: Importing libraries and initial setup.
2. Load Data: Loading and describing the dataset.
3. Exploratory Data Analysis (EDA): Descriptive analysis and visualization of the dataset features.
4. Feature Engineering: Transformations and encodings necessary to prepare the data for modeling.
5. Model Training: Training various classification models, including logistic regression, KNN, decision trees, random forests, and more.
6. Model Evaluation: Evaluating models using metrics such as accuracy, confusion matrix, and ROC curves.
7. Generating Predictions: Generating predictions for the test dataset and saving the results in CSV files.

Models Used

• Logistic Regression
• K-Nearest Neighbors (KNN)
• Decision Trees
• Random Forests
• AdaBoost
• XGBoost

How to Run the Project

1. Clone the repository.
2. Install the necessary dependencies.
3. Run the notebook 2410_kaggle_bank_campaingMarcelo.ipynb to reproduce the analysis and results.

Results

The prediction results can be found in the submissions/ folder, where each CSV file corresponds to the predictions generated by a specific model.

License

This project is licensed under the MIT License.