mrvutuananh / helpmate_ai Goto Github PK

HELPMATE_AI Semantic Search System

INTRODUCTION
------------
HELPMATE_AI is a sophisticated semantic search system designed to efficiently retrieve and rank documents based on the Retrieval-Augmented Generation (RAG) pipeline. This system leverages advanced NLP techniques to process PDF documents and provide relevant information in response to user queries.

FEATURES
--------
- Semantic search using RAG pipeline for precise document retrieval.
- PDF text and table extraction with vector representation generation.
- Cache implementation to enhance retrieval performance.
- Vector storage with ChromaDB for scalable search capabilities.

INSTALLATION
------------
1. Clone the repository:
   git clone https://github.com/MrVuTuanAnh/HELPMATE_AI.git
2. Navigate to the project directory:
   cd HELPMATE_AI
3. Install the required dependencies:
   pip install -r requirements.txt

REPORT:
------
- Short report:  HELPMATE_AI_Report.pdf
- Screen report: Screenshort.pdf
- Full & detail of Helpmate_ai : main.ipynb

USAGE
-----
- Use main.ipynb on jupyter for executing the project code.
- Follow the notebook instructions to perform document processing and semantic searches.

CONFIGURATION
-------------
- Set up environment variables as needed for database and API connections.
- Configure cache settings in the ChromaDB settings file.

CONTACT
-------
Project Link: https://github.com/MrVuTuanAnh/HELPMATE_AI

ACKNOWLEDGEMENTS
----------------
- OpenAI for language model APIs.
- Hugging Face for transformer models.
- The pdfplumber, tiktoken, and ChromaDB libraries.

Please ensure you follow the instructions for installation and setup closely, and refer to individual scripts for additional documentation on function usage.