DocumentGPT: An Innovative Research Tool
DocumentGPT is a cutting-edge web application designed to enhance the way individuals engage with their research documents. By leveraging OpenAI's Chat API, DocumentGPT offers a unique platform that combines the power of AI conversation with semantic search capabilities through Vector Databases. This tool simplifies the interaction with complex documents, allowing users to explore and gain insights effortlessly.
Key Features
-
Document Interaction: Users can upload their PDF research documents directly into the web application. This feature allows them to seamlessly view and navigate through their content within the app.
-
AI Chatbot Integration: Powered by OpenAI's Chat API, DocumentGPT enables users to chat with an AI chatbot. This functionality allows for a conversational exploration of the research document, providing a dynamic way to extract information and understand the content.
-
Semantic Search: Utilizing vector databases, DocumentGPT performs semantic searches across the document to provide precise answers. This feature enhances users' ability to find pertinent information quickly and accurately.
-
Web Contextualization: Users can receive additional context and information-rich answers by searching the web for related topics directly within the app.
-
Multi-platform Search: The tool facilitates Google and YouTube searches, offering automated search suggestions for efficient information retrieval.
-
Source Verification: Each generated result or answer can be verified for its sources, ensuring the reliability of the information provided.
Demonstration Preview
For a visual demonstration of how DocumentGPT functions, users can visit the following Demo Preview.
Getting Started
To start using DocumentGPT, follow these steps:
- Clone the repository using the command:
git clone https://github.com/your-username/DocumentGPT.git. - Install the necessary dependencies with:
pip install -r requirements.txt. - Set up your OpenAI API key within the web application.
- Launch the application using:
streamlit run main.py. - Access the app in your browser at
http://localhost:8501.
Available Tools
DocumentGPT boasts a variety of tools, including:
- Vector Database Retrieval Tool: Efficiently finds answers using a vector database.
- Arxiv Tool: Searches for scientific articles on arxiv.org.
- Search Tool: Enables web searches for general articles.
- Summarization Tool: Summarizes the content of entire documents as requested.
Future Enhancements
The project has room for potential improvements:
- AI Alignment: Enhancing model behavior through advanced prompt engineering techniques.
- Tool Expansion: Developing and integrating more tools for improved responses.
Deployment on Streamlit Cloud
Users can test the chatbot on Streamlit Cloud through this Deployed App. Please note that PDF display may have limitations based on the browser used, with optimal performance observed in Safari and Firefox.
Contribution and License
Contributions to DocumentGPT are warmly welcomed. Users can provide feedback, suggestions, or report bugs by creating issues on the GitHub repository. Additionally, enhancements can be shared through pull requests. This project is open-source and licensed under the MIT License.
Acknowledgements
DocumentGPT was developed using Streamlit, OpenAI Chat API, Langchain, among other open-source libraries. Special thanks to the developers and contributors of these technologies for enabling the development of this project.