Skip to content

Latest commit

ย 

History

History
116 lines (82 loc) ยท 4.16 KB

File metadata and controls

116 lines (82 loc) ยท 4.16 KB

๐ŸŽฎ EduGami - AI-Powered Gamified Learning Platform

License: MIT Project Stage AI Powered

๐Ÿ’ก Rethinking Gamification for WsCube
Instead of just leaderboards and points, we focused on high-impact, innovative features that provide real-world learning experiences.

๐ŸŽฏ Introducing Two Marketplaces

1๏ธโƒฃ Freelancing Simulator ๐Ÿ—๏ธ A marketplace like Upwork where students bid on dummy projects. Winners work with an AI avatar acting as their client to simulate real freelancing gigs.

2๏ธโƒฃ Job Simulation Hub ๐Ÿ’ผ Students apply for dummy job postings with a resume. They go through AI-powered hiring rounds (GD, interviews, etc.). Once hired, they join a team and work on projects collaboratively.

Machine Learning Model Description:


Script 1: Bid Ranking Prediction

Predicts freelancing bid ranks using a deep learning model for text clarity and a regression model for numerical features (experience, price, reputation). Combines both models to output a bid rank.


Script 2: Student Learning Style Clustering

Clusters students into learning styles (e.g., visual, fast, slow learners) based on behavior data (clicks, time spent, scores). Uses K-Means clustering, PCA for visualization, and provides personalized recommendations.


Script 3: Topic Recommendation with BKT and Collaborative Filtering

Recommends quiz topics by tracking skill mastery using Bayesian Knowledge Tracing (BKT) and collaborative filtering (SVD). Adjusts recommendations based on performance and time taken, focusing on weak areas.

๐Ÿ“– Table of Contents


๐Ÿš€ Problem Statement

Modern education systems struggle with:

  • โŒ Low student engagement
  • ๐Ÿงฉ One-size-fits-all learning approaches
  • ๐ŸŒ‰ Gap between theoretical knowledge and real-world application
  • โณ Lack of 24/7 personalized support

๐Ÿ’ก Our Solution

Next-Gen Gamified Learning Ecosystem

+ Combines game mechanics with AI personalization
+ Virtual career simulations with real-world relevance
+ Dynamic difficulty adjustment using reinforcement learning

System Architecture

๐Ÿ› ๏ธ Technical Stack

Category Technologies
AI/ML Core Python โ€ข TensorFlow โ€ข PyTorch โ€ข RLlib โ€ข Hugging Face Transformers
Generative AI GPT-4 โ€ข Claude 3 โ€ข Llama 3 โ€ข LangChain โ€ข Pinecone (Vector DB)
Backend Node.js โ€ข FastAPI โ€ข GraphQL โ€ข WebSocket
Database PostgreSQL โ€ข MongoDB โ€ข Redis โ€ข Elasticsearch
Frontend React โ€ข Three.js โ€ข Unity โ€ข WebGL
DevOps AWS โ€ข Docker โ€ข Kubernetes โ€ข GitHub Actions
Analytics Mixpanel โ€ข MLflow โ€ข Power BI

๐ŸŽฏ Key Features

Core Gamification

Feature Status Description
๐Ÿ† Dynamic Leaderboards โœ… Done Real-time ranking with Redis sorted sets
๐Ÿง‰ XP Token System ๐Ÿšง WIP Blockchain-based reward tracking
๐Ÿ… Skill Badges โœ… Done 50+ unlockable achievements

AI Innovations

Feature Tech Used Impact
๐Ÿค– AI Interview Coach GPT-4 + Unity Avatars 90% interview simulation accuracy
๐Ÿง  Adaptive Learning RLlib + K-Means 40% faster skill acquisition
๐Ÿ’ฌ 24/7 Tutor Bot RASA + LangChain 2s avg response time

Career Simulators

graph LR A[Learning Path] --> B{Level 5} B --> C[Freelance Marketplace] B --> D[Corporate Simulator] C --> E[Project Bidding] D --> F[Team Projects]

๐Ÿ“„ License

This project is licensed under the MIT License - see the LICENSE file for details.