Retreival-Augmented-Generation-RAG

Python-based FAQ retrieval system designed to answer common questions about Golden Retrievers

Golden Retriever FAQ Retrieval System

This repository contains a Python-based FAQ retrieval system designed to answer common questions about Golden Retrievers. It utilizes fuzzy matching and semantic search techniques to handle various ways a question may be phrased, offering relevant answers from a predefined FAQ database.

Features

• Comprehensive FAQ Dataset: The system is equipped with a wide range of AI generated FAQs about Golden Retrievers.
• Fuzzy Matching: Handles user queries that are close but not identical to predefined FAQ questions using fuzzy matching with the rapidfuzz library.
• Semantic Search: Uses embedding-based similarity (via language models) to find semantically similar questions, ensuring flexibility in understanding different phrasings of the same question.

Usage

1. Run the Jupyter Notebook:

   jupyter notebook

2. Load the FAQ Retrieval System: The FAQ retrieval system can be run directly in the notebook. You can query it by providing a natural language question related to Golden Retrievers.

   Example of a query:

   query = "How long do Golden Retrievers live?"
   answer = retrieve_with_fuzzy_faq(query)
   print(answer)

3. Customize the FAQ Database: The FAQ database can be easily modified by adding or removing questions in the faq dictionary. You can extend the system with new questions and answers about Golden Retrievers as needed.

4. Open a pull request.

You are welcome contributions in the form of additional FAQs, optimizations to the retrieval system, or improved handling of queries.

License

This project is licensed under the MIT License. See the LICENSE file for more details.