This project predicts student exam performance using Machine Learning and provides personalized study recommendations using Google's Gemini AI.
The system analyzes student study habits and academic factors to estimate exam scores and generate customized improvement plans.
- Predict student exam scores using Random Forest Regression
- Handle missing values and preprocess educational data
- Perform exploratory data analysis (EDA)
- Hyperparameter tuning using GridSearchCV
- Cross-validation for model evaluation
- Save trained model using Joblib
- Generate AI-powered study recommendations using Gemini
- Interactive Streamlit web application
- Personalized study schedules and exam preparation tips
Dataset: StudentPerformanceFactors.csv
- Hours Studied
- Attendance
- Previous Scores
- Sleep Hours
- Exam Score
- Python
- Pandas
- NumPy
- Matplotlib
- Seaborn
- Scikit-Learn
- Streamlit
- Google Gemini API
- Joblib
- Data Loading
- Data Cleaning
- Missing Value Handling
- Label Encoding
- Exploratory Data Analysis
- Feature Selection
- Model Training
- Hyperparameter Tuning
- Model Evaluation
- Model Deployment
The project uses Random Forest Regression to predict exam scores based on student academic factors.
- R² Score
- Cross Validation Score
git clone https://github.com/yourusername/student-performance-predictor-ai.git
cd student-performance-predictor-ai
pip install -r requirements.txtpython studentperformancefactors.pystreamlit run app.pyInput:
- Hours Studied: 5
- Attendance: 75%
- Previous Score: 65
- Sleep Hours: 6
Output:
- Predicted Exam Score
- AI Generated Study Plan
- Daily Study Schedule
- Revision Strategy
- Exam Preparation Advice
- Include all dataset features
- Compare multiple regression models
- Feature importance visualization
- Model explainability using SHAP
- Deployment on Streamlit Cloud
- Student performance dashboard
Abhinav Krishna C S
This project is for educational and academic purposes.