vanshilpatel / chatbot-mined-24

A repo for submitting the solution to the problem statement by Cactus Communications

Chat with any Scientific Documents

Introduction

Experience seamless document interaction with 'Chat with any Scientific Documents.' Upload PDFs, DOCX, LaTeX, HTML and Image files, press 'stream,' and effortlessly engage in a conversational interface. Pose questions related to your documents, and receive instant, contextually relevant answers, streamlining your workflow for efficient knowledge extraction.

Workflow

1. Document Loading: Load various document types, including PDFs, DOCX, LaTeX, and HTML, and Image into the application. Extract text content from the loaded documents.

2. Text Chunking: Divide the extracted text into smaller, manageable chunks for effective processing.

3. Language Model: Utilize a language model to create vector representations (embeddings) for each text chunk.

4. Similarity Matching: When a user poses a question, compare it with the vectorized text chunks to identify the most semantically similar ones.

5. Response Generation: Pass the selected text chunks to the language model for generating a response based on the relevant content extracted from the loaded documents.

6. User Interaction: Engage in a conversational interface, asking natural language questions related to the uploaded documents.

7. Receive Responses: Receive responses generated by the system, offering contextually relevant information from the processed documents.

Demo Video

Click here to watch the demo video

Dependencies and Installation

Install the required dependencies by running the following command:

pip install -r requirements.txt

Usage

To use the Chat with Scientific Document App, follow these steps:

1. Ensure that you have installed the required dependencies and added the OpenAI API key to the .env file.

2. Run the main.py file using the Streamlit CLI. Execute the following command:

   streamlit run main.py

The application will launch in your default web browser, displaying the user interface. Load multiple documents (you can upload documents like PDFs, DOCX, HTML files, LaTeX files, etc.) into the app by following the provided instructions. Ask questions in natural language about the loaded documents using the chat interface.