This project analyzes and predicts student performance in Mathematics based on socio-economic and academic factors using Machine Learning techniques and deploys it using Flask, AWS, and GitHub Actions for CI/CD.
✅ Understanding the Problem
📊 Data Collection & Checks
📈 Exploratory Data Analysis (EDA)
🛠 Data Pre-Processing
🤖 Model Training & Evaluation
🏆 Choosing the Best Model
🚀 Deployment with Flask & AWS
Understanding how student performance (test scores) is influenced by:
🧑🎓 Gender
🌍 Ethnicity
🎓 Parental Education
🍽 Lunch Type
📚 Test Preparation Course
📌 Categorical Features:
- Gender, Ethnicity, Parental Education, Lunch Type, Test Prep Course
📌 Numerical Features:
- Math Score, Reading Score, Writing Score
1️⃣ Data Collection & Preprocessing – Handling missing values, duplicates, outliers
2️⃣ EDA – Visualizing trends & correlations 📊
3️⃣ Feature Engineering – Encoding categorical data, scaling, splitting datasets
4️⃣ Model Training – Experimenting with multiple ML models
5️⃣ Model Selection – Comparing models using evaluation metrics
6️⃣ Web Interface Development – Flask-based UI for easy predictions
7️⃣ Deployment – Hosted on AWS EC2 with ECR & GitHub Actions for CI/CD
🐍 Python (Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn)
📓 Jupyter Notebook – Model training & analysis
🤖 Machine Learning (LinearRegression, RandomForestRegressor, GradientBoostingRegressor, KNeighborsRegressor, XGBRegressor, CatBoostRegressor, AdaBoostRegressor etc.)
🌍 Flask – Web interface for user-friendly predictions
☁ AWS (EC2, ECR) – Cloud deployment & containerized infrastructure
⚡ GitHub Actions – CI/CD automation for deployment
📌 Key insights into factors affecting student performance
📌 Accurate ML model for predicting student math scores
📌 Web app for real-time score prediction
📌 Cloud-deployed solution for accessibility
📧 Shiva Prasad Naroju - [email protected]
Let me know if you need any further improvements! 🚀🔥