Capstone Project

📌 Project Requirements

Your project must be a classical machine learning project using real-world data (not pre-solved Kaggle competitions or overused tutorial datasets like Titanic/MNIST). Apply techniques learned in the course: regression, classification, clustering, or deep learning with proper methodology including data cleaning, exploratory analysis, model selection, and evaluation.

⚠️ Academic Integrity

Plagiarism will result in immediate failure. All projects are individual - no code sharing between students. You must cite all external code sources, list AI tools used (ChatGPT, Copilot) in the appendix, properly reference your dataset origin, and write the report in your own words.

💡 What Makes a Good Project

Choose a problem with real impact: predict housing prices with feature engineering, classify customer churn, analyze financial sentiment, forecast energy consumption, or tackle a domain-specific classification problem. Your project should demonstrate both technical skill and practical insight.

❌ What to Avoid

Do not submit projects using overused datasets (Titanic, Iris, MNIST without innovation), Kaggle competitions with public solutions, purely theoretical work without implementation, or direct replications of course exercises. Projects must go beyond basic statistics to include genuine machine learning techniques.