EduRAG: Intelligent Teaching Assistant

🚀 Overview

EduRAG is a Retrieval-Augmented Generation (RAG) based AI system designed to generate accurate, context-aware answers from structured educational content.
It improves traditional question-answering by combining semantic search with LLM-based response generation.

---

🔥 Key Features

• Retrieval-Augmented Generation (RAG) pipeline for improved answer accuracy
• Custom chunking and chunk-merging strategy to enhance context quality
• Embedding-based semantic search for relevant content retrieval
• Multi-file JSON data processing and structuring
• Context-aware response generation with reduced noise

---

🧠 How It Works

1. Raw data is preprocessed and divided into chunks
2. Chunks are intelligently merged to improve context
3. Embeddings are created for semantic understanding
4. Top-k relevant chunks are retrieved based on query
5. LLM generates a final answer using retrieved context

---

🛠️ Tech Stack

• Python
• JSON Data Processing
• Embedding-based Semantic Search
• Retrieval-Augmented Generation (RAG)

---

📂 Project Structure

. ├── merge_chunks.py ├── preprocess_json.py ├── processing_query.py ├── mp3_to_json.py ├── video_to_mp3.py ├── README.md

---

🎯 Key Highlights

• Designed a custom chunk-merging mechanism to reduce context fragmentation
• Improved answer quality by optimizing chunk size and grouping strategy
• Built modular scripts for scalable data preprocessing and retrieval
• Focused on enhancing LLM performance through better context handling

---

⚡ Usage / Workflow

• Convert video content to audio using video_to_mp3.py
• Transcribe audio to structured JSON using mp3_to_json.py
• Preprocess raw data using preprocess_json.py
• Merge chunks for improved context using merge_chunks.py
• Perform query processing and retrieval using processing_query.py