🚀 Live Demo - Try it now!
An intelligent, GUI-based machine learning application that predicts student placement outcomes and generates personalized career roadmaps based on academic performance, interests, and skill levels.
✅ Placement Prediction using advanced machine learning models
✅ Model Comparison: Logistic Regression vs Random Forest
✅ Performance Metrics: Accuracy, F1 Score, Precision, Recall, ROC AUC
✅ Interactive GUI built with Tkinter
✅ Dynamic Career Roadmap Generator based on branch & interests
✅ Curated Learning Resources with links to top courses (Coursera, Udemy, etc.)
✅ Data Visualization with confusion matrices and ROC curves
- Python 3.x
- Tkinter (for GUI)
- Pandas, NumPy (data manipulation)
- Scikit-learn (ML models & preprocessing)
- Matplotlib, Seaborn (visualization)
- imbalanced-learn (SMOTE for class balancing)
-
Clone the repository:
git clone https://github.com/Shrey-003/Student_Placement_Predictor.git cd Student_Placement_Predictor -
Install dependencies:
pip install -r requirements.txt
-
Run the application:
python placement_gui.py
-
Clone and install (same as above)
-
Run Streamlit app:
streamlit run app.py
-
Or visit the live demo: https://studentplacementpredictor-cldvjgdxa8wlw2yulspj3b.streamlit.app
This project is deployed on Streamlit Cloud:
- URL: https://studentplacementpredictor-cldvjgdxa8wlw2yulspj3b.streamlit.app
- Status: ✅ Active
- Auto-updates: Pushes to main branch automatically redeploy
See STREAMLIT_DEPLOY.md for detailed deployment instructions.
- Launch the GUI by running
placement_gui.py - Enter student details:
- CGPA
- Number of Internships
- Number of Projects
- Workshops/Certifications
- SSC Marks
- HSC Marks
- Click "Predict Placement" to see the placement probability
- Get personalized roadmap:
- Select your academic branch
- Choose your technical interest
- Select your skill level (1-10)
- Click "Get Learning Resources" for a curated roadmap
-
Data Preprocessing
- Feature scaling with StandardScaler
- Custom feature weighting based on importance
- Polynomial feature transformation (degree=2)
- SMOTE for handling class imbalance
-
Model Training
- Logistic Regression (liblinear solver)
- Random Forest Classifier (100 estimators)
- Automatic selection of best-performing model
-
Evaluation Metrics
- Accuracy Score
- F1 Score
- Precision & Recall
- ROC AUC Score
- Confusion Matrix Visualization
Student_Placement_Predictor/
│
├── placement_model.py # ML model training & evaluation
├── placement_gui.py # Tkinter GUI application
├── roadmap_generator.py # Career roadmap generation logic
├── updated_placement_data.csv # Training dataset
├── requirements.txt # Python dependencies
├── screenshot.png # Application screenshot
└── README.md # Project documentation
The roadmap generator provides personalized learning paths based on:
- Academic Branch (CSE, ECE, ME, Civil, etc.)
- Interest Area (Data Science, Software Dev, AI/ML, Finance, etc.)
- Skill Level (1-10 scale)
Output includes:
- Recommended online courses
- Industry certifications
- Tools & technologies to master
- Step-by-step learning roadmap
The system automatically selects the best-performing model:
- Random Forest typically achieves ~85-90% accuracy
- Logistic Regression provides interpretable results
- Real-time metrics displayed in the GUI
Contributions are welcome! Feel free to:
- Report bugs
- Suggest new features
- Submit pull requests
This project is open-source and available for educational purposes.
Shrey Patel
GitHub: @Shrey-003
- Dataset curated from academic placement records
- Career roadmaps based on industry best practices
- Course recommendations from leading platforms