🎓 Student Performance Predictor (Machine Learning Project)

A complete end-to-end Machine Learning project that predicts a student's Performance Index based on academic and lifestyle factors using Linear Regression.

---

🚀 Project Overview

This project analyzes how different factors affect student performance and builds a predictive model using real-world-like data.

It takes user inputs such as study hours, sleep, and previous scores, and predicts the expected performance index.

---

📊 Dataset Features

The dataset contains 10,000 entries with the following features:

• 📘 Hours Studied (Numeric)
• 📝 Previous Scores (Numeric)
• 🎯 Extracurricular Activities (Categorical: Yes/No)
• 😴 Sleep Hours (Numeric)
• 📄 Sample Question Papers Practiced (Numeric)
• 🎯 Performance Index (Target Variable)

---

🧠 Machine Learning Workflow

✅ Steps Performed:

1. Data Loading
2. Data Cleaning
3. Encoding Categorical Variables
4. Feature Scaling (StandardScaler)
5. Train-Test Split
6. Model Training (Linear Regression)
7. Model Evaluation
8. User Input Prediction System

---

📈 Model Performance

• 🔹 Mean Absolute Error (MAE): 1.61
• 🔹 R² Score: 0.9889

👉 The model performs extremely well, explaining ~99% variance in the data.

---

🖥️ How It Works

After training, the program allows real-time predictions:

Example:

Enter student details: Hours Studied: 5 Previous Scores: 85 Extracurricular Activities (Yes/No): yes Sleep Hours: 8 Sample Papers Practiced: 0

Predicted Performance Index: 71.21

---

⚠️ Note

You may see this warning:

UserWarning: X does not have valid feature names

👉 This happens because the model was trained with feature names but prediction input is a raw array.

✅ Fix (Optional Improvement):

Use a DataFrame for prediction instead of a list.

---

🛠️ Tech Stack

• 🐍 Python
• 📊 NumPy
• 📁 Pandas
• 🤖 Scikit-learn

---

📂 Project Structure

Student_Performance_Predictor/ │── main.py │── Student_Performance.csv │── README.md

---

🔮 Future Improvements

• 🔹 Add Polynomial Regression
• 🔹 Try Advanced Models (Random Forest, XGBoost)
• 🔹 Hyperparameter Tuning
• 🔹 Build Web App (Streamlit)

---

💡 Key Learnings

• Real-world ML pipeline
• Feature scaling importance
• Handling categorical data
• Model evaluation metrics
• Building interactive ML systems

---

⭐ Final Thoughts

This project is a strong beginner-to-intermediate ML project that demonstrates:

✔ End-to-end pipeline
✔ Real-world data handling
✔ Model deployment logic (CLI-based)

---

🚀 Built with dedication by Aeindri