DocuAnswer is a powerful document-based question-answering system that leverages NLP techniques to extract and provide answers from given pdf documents.
- format support: Easily handle PDFs
- Natural Language Processing (NLP): Uses LangChain, Transformers, FAISS and ChatGPT to answer questions
- Interactive interface: Streamlit interface to interact with the Application.
To install and run DocuAnswer locally, follow these steps:
-
Clone this repository:
git clone https://github.com/mittapallynitin/DocuAnswer.git cd DocuAnswer -
Install the required dependencies:
pip install -r requirements.txt
-
Run the application:
streamlit run main.py
- Python
- Langchain for NLP processing and Chunking Documents
- PyPDF to parse PDF files
- SentenceTransformer for creating embeddings
- FAISS for storing and retrieving relevant document chunks
- Flask to provide an REST API endpoint
- Streamlit to provide interactive web interface
- Upload a document through the web interface.
- Ask a question in natural language.
- Receive the answer extracted from the document.
Contributions are welcome! To get started:
- Fork the repository.
- Create a new branch.
- Make your changes.
- Submit a pull request.
For questions, suggestions, or collaboration opportunities, feel free to reach out via LinkedIn or email me at mittapally.nitin@gmail.com.