Harnessing Behavioral Analytics for Student Wellbeing
Team Name: ELITERS
| Name | Role |
|---|---|
| Savanth G (Team Lead) | AI/ML Development (Model Design, Recommendation Engine & Risk Scoring) |
| S MOHAMMAD PARDEEN | Backend & Frontend Integration, API Development |
| Vikas GJ | Front-End Development |
| N NAUMAN PASHA | Data Processing & Data Management |
Burnout among students is not a sudden event — it is a gradual decline driven by sustained academic pressure, irregular habits, and unnoticed stress patterns. Most existing systems respond only after the impact is visible.
MindGuard AI shifts this approach from reaction to prevention.
It leverages behavioral data and machine learning to identify early signs of burnout and provide timely, personalized interventions — enabling students to act before their performance and wellbeing are affected.
Students today face increasing academic demands, yet there are no reliable systems to:
- Detect early indicators of burnout
- Monitor behavioral patterns meaningfully
- Provide proactive, data-driven guidance
As a result, interventions are delayed, leading to reduced academic performance and declining mental wellbeing.
MindGuard AI introduces a structured system for early burnout detection by combining behavioral tracking with intelligent analysis.
The system focuses on key inputs:
- Sleep patterns
- Study duration
- Screen time
- Mood indicators
These are analyzed to generate a dynamic burnout risk score, along with actionable recommendations tailored to each user.
-
Data Collection
Users input daily behavioral metrics such as sleep, study time, screen usage, and mood. -
Pattern Analysis
A hybrid model evaluates trends and detects stress-related deviations. -
Risk Scoring
A burnout risk score (0–100) is computed in real time. -
Personalized Feedback
Users receive targeted recommendations to reduce burnout risk.
The current system includes:
- Behavioral data input interface
- Burnout risk scoring system
- Core analytical model for pattern detection
- Personalized recommendation engine
This implementation focuses on delivering a usable, student-centric solution with immediate value.
Planned extensions (not yet implemented):
- Institutional dashboard for aggregated insights
- Identification of at-risk student groups
- Peak stress period analysis
- Integration with campus wellness systems
These features aim to expand the platform into a broader decision-support system.
MindGuard AI uses a hybrid approach:
- Behavioral Trend Analysis to track gradual changes
- Machine Learning Models for predictive scoring
- Adaptive Learning to improve accuracy over time
- HTML5
- CSS3 / Tailwind CSS
- JavaScript
- Python
- fastapi
- Scikit-learn
- NumPy
- Pandas
- Matplotlib
- VS Code
- Git & GitHub
MindGuard AI follows a modular, layered architecture ensuring scalability and separation of concerns.
- User Interface: Handles input and displays insights
- API Layer: Connects frontend and backend logic
- Data Input Module: Processes user behavioral data
- ML Processing Engine: Generates burnout risk score
- Recommendation Engine: Produces actionable suggestions
- Data Storage: Maintains user data for analysis
- Enables early burnout detection
- Improves student wellbeing and performance
- Supports data-driven decision making
- Provides a foundation for scalable mental health systems
To build a scalable AI-driven ecosystem for student wellbeing that integrates:
- Real-time behavioral monitoring
- Advanced predictive analytics
- Institution-level insights
The goal is to enable early intervention and resilient learning environments.
##States
ONGOING
MindGuard AI addresses a critical gap by introducing a proactive, data-driven approach to burnout detection. Instead of reacting to outcomes, it empowers users to take control early — transforming how burnout is identified and prevented.
Savanth G
📧 savanthg14@gmail.com
🌐 LinkedIn : linkedin.com/in/savanth-g-65454a36b