MasteryPath – AI-Powered Adaptive Learning and Skill Assessment Platform

MasteryPath is an AI-powered adaptive learning and skill assessment platform designed to make programming education more personalized and effective. Unlike traditional learning platforms that provide the same questions and practice exercises to every student, MasteryPath adapts the learning experience based on each student's knowledge, performance, and understanding.

The platform focuses on Data Structures and Algorithms using Python and evaluates individual programming skills such as Arrays, Linked Lists, and other fundamental concepts. It continuously analyzes student responses to estimate how well each concept has been understood and identifies areas where additional practice is required.

At the heart of MasteryPath is Bayesian Knowledge Tracing (BKT), a probabilistic machine learning technique used to estimate a student's knowledge state. It considers four important parameters: prior knowledge, learning probability, guess probability, and slip probability. Based on these parameters and student responses, the system updates the estimated mastery probability for each skill.

MasteryPath uses an adaptive question selection mechanism to identify weaker unlocked skills and recommend relevant practice questions. When a student's estimated mastery reaches the configured threshold of 95%, the corresponding skill is considered mastered, helping students progress through their learning journey systematically.

The platform provides dedicated experiences for students and teachers. Students can access personalized practice sessions, monitor skill-wise mastery levels, and visualize their learning progress through interactive dashboards. Teachers can monitor classroom performance, analyze student progress through skill-based heatmaps, and identify concepts where students need additional support.

MasteryPath also includes a Model Lab that provides insights into Bayesian Knowledge Tracing parameters and supports model fitting using Expectation-Maximisation (EM). Its evaluation module helps assess model performance using metrics such as AUC and compare adaptive learning approaches against baseline methods.

Developed using React, TypeScript, Python, FastAPI, SQLite, and machine learning technologies, MasteryPath demonstrates how artificial intelligence and probabilistic modelling can improve personalized education.

This project is suitable for students interested in Artificial Intelligence, Machine Learning, Python development, full-stack application development, educational technology, and intelligent recommendation systems.