Student Digital Twin System
A simulation-based academic and career analytics system that models how daily student habits influence long-term academic performance and placement readiness. The project focuses on explainable logic-based modeling instead of black-box machine learning.
Back to ProjectsProject Overview
The Student Digital Twin models a student using behavioral and academic inputs such as attendance, current marks, daily study hours, sleep hours, skill level, and internship effort. It simulates academic progress over time and generates final marks prediction, career readiness score, placement probability, scenario comparison, explainable impact breakdown, and primary growth driver analysis.
Core Idea
Traditional academic systems evaluate only past marks and attendance. This system answers a more practical question about how today's habits affect future academic and career outcomes.
It uses time-lag simulation modeling, weighted contribution logic, scenario-based improvement comparison, and explainable factor-level impact analysis.
Tech Stack
- Python
- Streamlit
- Logic-based Simulation
- Academic Analytics
- Git & GitHub
How It Works
The system takes student inputs, processes them through explainable logic, and updates the projected academic and career outcomes over time. Each simulation step shows how changes in habit patterns can improve or weaken the final result.
Challenges Faced
The main challenge was keeping the model explainable while still making it useful for realistic long-term prediction. Another challenge was balancing different student factors so the output remained meaningful and easy to understand.
What I Learned
This project taught me how to structure a simulation around real-world behavior and keep the logic easy to explain. It also improved my understanding of building systems that are practical, transparent, and centered on user learning.
Latest Update
I am continuing to refine the simulation model and expand the analytics so the system becomes more useful for students over time.