Emotion-Aware Learning Platform

Your Personal Path to Smarter Learning

Comment

Inspiration

Digital courses often lose learners to boredom, confusion, or stress. MoodTutor tackles this by meeting students where they are emotionally, and by giving teachers visibility into those patterns so they can fix tough spots fast. The goal: turn passive learning into an engaging, emotion-adaptive journey.

What it does

  • Emotion-Adaptive Learning: Students choose among Happy, Bored, Confused, Stressed at checkpoints; content adapts instantly (quizzes/games for Bored, hints/simpler steps for Confused, break/relax prompts for Stressed, next-level content for Happy).
  • Gamification: Points per step, lesson-completion bonus, engagement score (0–100), progress bars, and badges groundwork.
  • Teacher Analytics: Emotion distribution per lesson, engagement metrics, active lessons, and insights to identify tricky sections.
  • Multi-Role Access: Secure JWT auth with student/teacher roles. Sample Lessons: Python & Math, multi-step with emotion checkpoints and mixed content (quizzes, games, hints, videos, breaks).

How we built it

  • Backend: FastAPI, MongoDB, JWT auth, bcrypt hashing, CORS.
  • 15+ REST endpoints for users, lessons, progress, emotion logs.
  • Frontend: React , responsive design, animations.
  • Auth via React Context; local state for lessons; clean loading/error states.
  • Dashboards with clear charts for emotion breakdowns and engagement.

Challenges we ran into

  • API contract mismatches fixed for emotion recording and progress updates.
  • Getting engagement scoring to feel fair across diverse emotion sequences.
  • Designing role-based routes and protecting teacher endpoints.
  • Data modeling for high-volume emotion entries while keeping queries snappy.
  • Nailing mobile responsiveness and accessible controls for quick emotion input.

Accomplishments that we’re proud of

  • End-to-end emotion-adaptive flow works smoothly across lessons.
  • Teacher Analytics provides actionable insights, not just charts.
  • 95.8% backend test pass rate (23/24) and all core front-end flows verified.
  • Modern, accessible UI with consistent branding, ARIA, keyboard navigation.
  • Robust JWT multi-role auth and persistent MongoDB storage.

What we learned

  • Small, well-defined emotion states are enough to drive meaningful adaptation.
  • Clear API specs + validation prevent most integration bugs.
  • Micro-rewards (points, progress, badges) materially boost engagement.
  • Teachers want why content is hard (emotion context), not just completion stats.
  • Accessibility and responsive design pay off in real usage.

What’s next for Emotion-Aware Learning Platform

  • Optional camera-based emotion detection (on-device) with strict privacy controls.
  • Adaptive difficulty & mastery paths using knowledge tracing.
  • LLM-generated hints/explanations tailored to emotion + progress.
  • Richer gamification: badges, streaks, leaderboards, class challenges.
  • Mobile app + offline mode for low-connectivity environments.
  • Scalability & security: rate limiting, audit logs, multi-tenant setup, data retention policies.
Share this project:

Updates