ITC6103 – Applied Machine Learning

Institution: MSc in Data Science, the American College of Greece

Term: Winter 2024

Full Codebase and Report: GitHub Repository

Overview:

This project presents a comprehensive application of supervised and unsupervised machine learning techniques, including regression, classification, and clustering.

The team investigates the effectiveness of various models and preprocessing pipelines through dedicated Jupyter notebooks.

Project Structure

.
├── AML_Kapsalis_Stavrogiannis_Tzoras.docx   # Final written report of the project
├── classification_part.ipynb                # Classification models and evaluation
├── clustering_part.ipynb                    # Clustering methods and analysis
└── regression_part.ipynb                    # Regression models and performance

Component Descriptions

:AML_Kapsalis_Stavrogiannis_Tzoras.docx: A detailed report outlining the project background, data preprocessing, methodology, results, and conclusions.
classification_part.ipynb: Implements supervised classification techniques such as logistic regression, decision trees, and possibly ensemble methods. Includes model evaluation metrics like accuracy, precision, recall, and confusion matrices.
clustering_part.ipynb: Applies unsupervised learning to discover structure in the data. Techniques likely include k-means, DBSCAN, or hierarchical clustering, with accompanying visualizations.
regression_part.ipynb: Focuses on predicting continuous outcomes using linear regression, regularized models, and error metrics such as MAE, MSE, and R².

Getting Started

To run the notebooks:

Install dependencies using your preferred environment manager.
Open each .ipynb file in Jupyter or VSCode.
Follow along with the cells; most are self-contained and annotated.

Notes

All notebooks are intended to be executed sequentially from top to bottom.
Ensure required datasets (if any) are provided or loaded from the correct paths.

← Back to projects