Assignment: Supervised Machine Learning (Group Project)
Objective
The objective of this assignment is to apply and compare multiple supervised machine
learning algorithms on a single dataset and evaluate their performance.
Students will gain practical experience using real-world datasets from Kaggle.
Group Allocation & Datasets
Each group will work on ONE dataset only:
Group 1: House Price Prediction (Regression)
• [Link]
techniques
Group 2: Customer Churn (Classification)
• [Link]
Group 3: Loan Prediction (Classification)
• [Link]
Group 4: Student Performance (Regression/Classification)
• [Link]
Group 5: Credit Card Fraud (Classification)
• [Link]
Group 6: Diabetes Prediction (Classification)
• [Link]
Group 7: Titanic Survival (Classification)
• [Link]
Group 8: Wine Quality (Regression/Classification)
• [Link]
Group 9: Heart Disease (Classification)
• [Link]
Group 10: Car Price Prediction (Regression)
• [Link]
1. Task Requirements
Each group must:
• Use their assigned dataset
• Apply at least 5 supervised machine learning algorithms
• Compare the performance of all models
2. Required Algorithms
You must select from the following:
• Linear Regression
• Logistic Regression
• Decision Tree
• Random Forest
• Support Vector Machine (SVM)
• K-Nearest Neighbors (KNN)
• Naive Bayes
• Gradient Boosting (e.g., XGBoost)
Choose algorithms appropriate to your dataset type:
• Regression → Linear Regression, Random Forest Regressor, etc.
• Classification → Logistic Regression, SVM, etc.
3. Data Preprocessing
You must perform:
• Data cleaning (handle missing values)
• Encoding categorical variables
• Feature scaling (if needed)
• Train-test split (70/30 or 80/20)
4. Implementation (Coding)
Use Python with:
• Pandas
• NumPy
• Scikit-learn
• Matplotlib
For EACH model:
• Train the model
• Make predictions
• Evaluate performance
5. Evaluation Metrics
Classification:
• Accuracy
• Precision
• Recall
• F1-score
• Confusion Matrix
Regression:
• MAE
• MSE
• R² Score
6. Model Comparison
This is the most important part:
• Compare all models in a table
• Identify the best-performing model
• Explain why it performed better
• Discuss overfitting/underfitting
7. Report Writing
Structure:
1. Introduction
2. Dataset Description
3. Data Preprocessing
4. Algorithms Used
5. Implementation
6. Results (tables + graphs)
7. Comparison & Discussion
8. Conclusion
Length: 5–10 pages
8. Code Submission
• Submit:
o .ipynb (Jupyter Notebook) or .py file
• Code must:
o Be well commented
o Be clearly structured
o Include all steps
9. Presentation
Each group will present for 10–15 minutes
Include:
• Dataset overview
• Algorithms used
• Results and graphs
• Model comparison
• Final conclusion
10. Evaluation Criteria
Criteria Marks
Data preprocessing 15
Model implementation 25
Model comparison 20
Code quality 10
Report quality 15
Presentation 15
Total 100
11. Submission Requirements
Each group must submit:
• Report (PDF/Word)
• Code (Notebook/Python file)
• Slides (PPT/PDF)
12. Important Notes
• Work in groups of 2–4 students
• Use original work (no plagiarism)
• Clearly explain your results
• Focus on understanding, not just coding