Inspiration
Manual exam management is slow, error-prone, and unfair. Spreadsheets don't scale. We used AI to automate student seating, room allocation, and faculty assignment.
What it does
Desktop app using K-Means clustering to automatically generate fair exam plans. Handles 2,500+ students across departments, batches, and shifts in seconds.
How we built it
- Python + Tkinter (GUI)
- scikit-learn (K-Means, StandardScaler)
- pandas, numpy (data)
- matplotlib (charts)
- openpyxl (Excel) Pipeline: Data → Encoding → Scaling → K-Means → Seating → Faculty → Reports ## Challenges
- Finding optimal domain/batch features for clustering
- Managing rooms with different capacities across shifts
- Matching faculty to all domains in each room
- Making interactive seat grid responsive ## Accomplishments
- Complete exam plan in one click
- Visual classroom with clickable seats
- AI insights with confidence scores (97%+)
- 4 export formats (PDF, Excel, CSV, TXT)
- Global search across all data ## What we learned
- K-Means groups students effectively with good features
- Encoding + scaling improves clustering dramatically
- UI/UX makes AI feel simple
- Real problems can be solved with accessible AI
What's next
- Web version (Flask/React)
- Automatic k selection (Elbow method)
- Role-based logins
- SIS integration
- Mobile app
- Email notifications
Log in or sign up for Devpost to join the conversation.