Machine (Supervised) Learning for Bankers

Preprocessing, Classification & Regression using Python

This repository contains the teaching materials, datasets, notebooks, and presentation slides for the course Machine (Supervised) Learning for Bankers.

The course introduces the complete supervised machine learning workflow using real-world banking datasets and focuses on practical implementation with Python and Scikit-Learn.

---

Course Objectives

After completing this course, students will be able to:

• Understand the supervised machine learning workflow.
• Perform data preprocessing on banking datasets.
• Handle missing values, duplicates, invalid values, and outliers.
• Encode categorical variables.
• Scale numerical variables.
• Handle class imbalance using undersampling and SMOTE.
• Train and evaluate classification models.
• Train and evaluate regression models.
• Compare different machine learning algorithms.

---

Repository Structure

.
├── Slides/
│   └── Machine (Supervised) Learning for Bankers.pdf
│
├── Notebooks/
│   ├── ML_Classification.ipynb
│   └── ML_Regression.ipynb
│
├── Datasets/
│   ├── Dataset_Risque_Credit.xlsx
│   └── bank_credit_dataset.csv
│
├── Figures/
│
├── README.md
└── LICENSE

---

Topics Covered

Part I — Data Preprocessing

• Data Exploration
• Missing Values
• Duplicate Records
• Invalid Values
• Outlier Detection
• One-Hot Encoding
• Feature Scaling
• Data Leakage
• Class Imbalance
  • Random Undersampling
  • SMOTE

---

Part II — Classification

Algorithms covered:

• Logistic Regression
• k-Nearest Neighbors (KNN)
• Decision Tree
• Support Vector Machine (SVM)

Topics include:

• Model intuition
• Hyperparameters
• Feature scaling
• Training
• Prediction
• Model evaluation

---

Part III — Regression

Algorithms covered:

• Linear Regression
• Ridge Regression
• Lasso Regression
• ElasticNet
• KNN Regressor
• Decision Tree Regressor
• Support Vector Regression (SVR)

Evaluation metrics:

• Mean Absolute Error (MAE)
• Mean Squared Error (MSE)
• Root Mean Squared Error (RMSE)
• Coefficient of Determination ($R^2$)

---

Datasets

Credit Scoring Dataset

Used for classification.

Target:

Credible

Task:

Predict whether a customer should be approved or rejected for a loan.

---

Credit Amount Dataset

Used for regression.

Target:

Credit Amount

Task:

Predict the amount of credit to grant to a customer.

---

Technologies

• Python
• Jupyter Notebook
• NumPy
• Pandas
• Matplotlib
• Scikit-Learn
• imbalanced-learn

---

Installation

Clone the repository

git clone https://github.com/yourusername/machine-learning-for-bankers.git

Move into the project

cd machine-learning-for-bankers

Create a virtual environment (optional)

python -m venv venv

Activate it

Windows

venv\Scripts\activate

Linux / macOS

source venv/bin/activate

Install dependencies

pip install -r requirements.txt

---

Running the Notebooks

Launch Jupyter Notebook

jupyter notebook

Open:

• ML_Classification.ipynb
• ML_Regression.ipynb

---

Learning Workflow

Load Dataset
      │
      ▼
Explore the Data
      │
      ▼
Preprocess the Data
      │
      ▼
Split Train/Test
      │
      ▼
Train the Model
      │
      ▼
Predict
      │
      ▼
Evaluate
      │
      ▼
Compare Models

---

Example Algorithms

Classification	Regression
Logistic Regression	Linear Regression
KNN	KNN Regressor
Decision Tree	Decision Tree Regressor
SVM	Support Vector Regression

---

Evaluation Metrics

Classification

• Accuracy
• Precision
• Recall
• F1-score
• ROC-AUC
• Confusion Matrix

---

Regression

• MAE
• MSE
• RMSE
• $R^2$

---

Intended Audience

This material is designed for:

• Banking professionals
• Data Analysts
• Data Scientists
• Master's students
• AI students
• Anyone interested in applied machine learning for finance

---

Prerequisites

Students should have basic knowledge of:

• Python programming
• Linear algebra
• Probability and statistics
• Basic SQL (recommended)

---

Author

Gradi Kamingu Lubwele

M.Sc.

Optimall

Department of Mathematics, Statistics and Computer Science

Faculty of Science and Technology

University of Kinshasa

---

Citation

If you use these teaching materials, please cite:

@misc{kamingu2026bankml,
  author = {Gradi Kamingu Lubwele},
  title = {Machine (Supervised) Learning for Bankers},
  year = {2026},
  publisher = {GitHub},
  url = {https://github.com/yourusername/machine-learning-for-bankers}
}

---

License

This project is released under the MIT License.

---

⭐ Support

If you find this repository useful:

• ⭐ Star the repository
• 🍴 Fork it
• 🐛 Report issues
• 💡 Suggest improvements
• 🤝 Contribute with pull requests

---

"Good data and sound preprocessing are the foundation of every successful machine learning model."