AILA - An Intelligent Lecturing Assistant

A research-driven lecturing assistant featuring LLM-powered knowledge extraction, automated slide segmentation, and interactive concept maps for a smarter instructor and student experience.

Comment

AILA – An Intelligent Lecturing Assistant

This repository contains the full code and research artifacts for the AILA project.
AILA (AI‑powered Intelligent Lecturing Assistant) is a research system that supports secure logins, lecture material upload, knowledge extraction, automated segmentation, retrieval‑augmented AI teaching assistant, and role‑based dashboards for instructors and students.

Project Structure

  • aila_backend/ – FastAPI + SQLite backend (lecture processing, knowledge graph, MCQs, quizzes)
  • aila_frontend/ – Next.js/React frontend (dashboards, concept map UI, quiz tools)

Key Features

  • Secure login/signup with instructor and student roles
  • Course, week, and lecture organization and upload
  • Slide/segment extraction, LLM‑powered summaries and MCQ generation
  • Knowledge‑graph construction for each course/week
  • Instructor week view centered on an expandable concept map:
    • Core concepts shown first
    • Click a concept to reveal related sub‑concepts
    • Details panel with summary, slide content and MCQ generators
  • Spaced retrieval practice and quiz analytics via interactive dashboards
  • Modular, research‑oriented codebase for rapid iteration

Getting Started

Backend

See aila_backend/README.md for detailed backend setup.

  1. cd aila_backend
  2. pip install -r requirements.txt
  3. uvicorn main:app --reload

The backend will create ailastar.db automatically and listens on process.env.NEXT_PUBLIC_API_URL || 'http://localhost:8000' by default.

Frontend

See aila_frontend/README.md for detailed frontend setup.

  1. cd aila_frontend
  2. npm install
  3. npm run dev

The frontend runs on http://localhost:3000 by default and expects the backend at process.env.NEXT_PUBLIC_API_URL || 'http://localhost:8000' (configurable via environment).

Development Notes

  • All API endpoints are scoped under /api/ on the backend for easy consumption from the frontend.
  • Uses local SQLite for quick prototyping; can be upgraded to Postgres or another RDBMS.
  • Environment variables (e.g., Google API key) must be provided via .env in the backend and frontend as needed.

Faculty: Prof. Yuan An, CCI, Drexel University
Student: Ruhma Hashmi, CCI, Drexel University
** Research Project**

Built With

Share this project:

Updates