Kinetic Modeling - RAG

My Research Journey

During my work in kinetic modeling as a Graduate Student Researcher at Dr. Wang Lab, I encountered a significant challenge: managing and quickly accessing information across multiple research documents. This tool is my solution to streamline the research process.

Project Overview

A custom RAG is designed to:

• Upload and index research PDFs
• Add relevant web resources
• Perform intelligent, cross-document searches
• Generate context-aware research insights

Technologies Used

• Web Framework: Streamlit
• Document Retrieval: LlamaIndex
• Database: MongoDB
• AI Capabilities: OpenAI's Language Model

Setup Instructions

Requirements

• Python 3.8+
• MongoDB Account
• OpenAI API Key (recommended)

Installation

1. Clone the repository

git clone https://github.com/AakashKotha/Kinetic-Modeling-RAG.git
cd Kinetic-Modeling-RAG

2. Create Virtual Environment

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install Dependencies

pip install -r requirements.txt

4. Configure Environment Create a .env file with:

OPENAI_API_KEY=your_openai_api_key
MONGO_URI=your_mongodb_connection_string
ADMIN_USERNAME=your_admin_username
ADMIN_PASSWORD=your_admin_password

Running the Application

streamlit run streamlit_app.py

Research Motivation

As a researcher, I was frustrated by:

• Time-consuming manual document searches
• Difficulty tracking research information
• Inefficient knowledge compilation

This tool automates and simplifies these challenges.