Amna2568/Research_Forge

🚀 ResearchForge - Setup & Installation Guide

📖 Overview

ResearchForge is an AI-powered research assistant that automates literature review, gap analysis, and hypothesis generation. This guide will help you get the project running on your local machine.

🎯 What This Project Does

ResearchForge performs a complete 4-phase research analysis pipeline:

1. 1.📄 Paper Collection - Automatically gathers 10-15 research papers from arXiv, Google Scholar, and ClinicalTrials.gov
2. 2.📚 Literature Review - Generates comprehensive academic review using GPT-3.5-turbo
3. 3.🔍 Gap Analysis - Identifies research gaps using dual AI models (Flan-T5-Large + GPT-4)
4. 4.💡 Hypothesis Generation - Creates actionable research proposals with novelty scoring

Time savings: What takes researchers weeks → Done in 2-3 minutes

⚙️ System Requirements

• —Python: 3.11 or higher
• —RAM: 16GB minimum (for Flan-T5-Large model)
• —Storage: 5GB free space (for AI models)
• —OS: Windows, macOS, or Linux
• —Internet: Required for API calls and model downloads

🛠️ Prerequisites

Before starting, you'll need:

1. OpenAI API Key (Required)

• —Sign up at: https://platform.openai.com/
• —Create API key at: https://platform.openai.com/api-keys
• —Cost: ~$0.42 per complete analysis

2. Google Custom Search API (Required)

• —Create project at: https://console.cloud.google.com/
• —Enable Custom Search API
• —Create credentials: https://developers.google.com/custom-search/v1/introduction
• —Create Custom Search Engine: https://programmablesearchengine.google.com/
• —Cost: Free tier (100 queries/day)

3. HuggingFace Token (Optional)

• —Sign up at: https://huggingface.co/
• —Get token at: https://huggingface.co/settings/tokens
• —Used for: Faster model downloads (optional)

📦 Installation Steps

Step 1: Clone the Repository

git clone https://github.com/Radia-987/ResearchForge.git
cd ResearchForge

Step 2: Create Virtual Environment

Windows:

python -m venv venv
.\venv\Scripts\activate

macOS/Linux:

python3 -m venv venv
source venv/bin/activate

Step 3: Install Dependencies

pip install -r requirements.txt

⏳ Note: First run will download ~3.5GB of AI models:

• —Flan-T5-Large (3.13 GB)
• —Sentence-BERT (80 MB)
• —This is a one-time download

Step 4: Configure API Keys

Create a .env file in the project root directory:

# Windows
notepad .env

# macOS/Linux
nano .env

Add your API keys:

OPENAI_API_KEY=sk-proj-xxxxxxxxxxxxxxxxxxxx
GOOGLE_API_KEY=AIzaSyxxxxxxxxxxxxxxxxxx
GOOGLE_SEARCH_ENGINE_ID=xxxxxxxxxxxxxxxxxxxx
HF_TOKEN=hf_xxxxxxxxxxxxxxxxxxxx

⚠️ Important: Never commit this file to Git (it's already in .gitignore)

Step 5: Verify Installation

Check that all dependencies are installed:

pip list

You should see:

• —streamlit==1.31.0
• —openai==1.12.0
• —transformers>=4.41.0
• —torch>=2.1.0
• —sentence-transformers>=5.0.0

▶️ Running the Application

Start the Streamlit App:

streamlit run app.py

Expected Output:

You can now view your Streamlit app in your browser.

Local URL: http://localhost:8501
Network URL: http://192.168.x.x:8501

Open in Browser:

The app will automatically open at http://localhost:8501

If it doesn't, manually navigate to the URL shown in the terminal.

🎨 Using the Application

1. 1.Enter Research Topic
2. 2.Example: "federated learning privacy"
3. 3.Example: "machine learning for drug discovery"

1. 1.Click "🚀 Start Complete Analysis"

1. 1.Wait 2-3 minutes (progress shown in terminal)

1. 1.View Results in 3 Tabs:
2. 2.📚 Literature Review
3. 3.🔍 Gap Analysis
4. 4.💡 Research Hypotheses
5. 5.🧪 Experiment Design

📊 What Gets Downloaded

On first run, the following models are automatically downloaded:

Total: ~3.2 GB (stored in ~/.cache/huggingface/)

🔑 API Costs

Estimated costs per complete analysis:

🧪 Testing the Setup

Quick Test:

python -c "from multi_agent_system import MultiAgentSystem; print('✅ Import successful!')"

Check OpenAI Connection:

python -c "from dotenv import load_dotenv; from openai import OpenAI; import os; load_dotenv(); client = OpenAI(api_key=os.getenv('OPENAI_API_KEY')); print('✅ OpenAI connected!')"

🔧 Troubleshooting

Issue: `ModuleNotFoundError: No module named 'streamlit'`

Solution:

pip install -r requirements.txt

Issue: `OPENAI_API_KEY not found`

Solution:

• —Ensure .env file exists in project root (same folder as app.py)
• —Check API key format: OPENAI_API_KEY=sk-proj-...
• —Restart the Streamlit app after creating .env

Issue: `httpx.ConnectError` or API connection fails

Solution:

pip uninstall httpx
pip install httpx==0.24.1

Issue: Model download fails or hangs

Solution:

pip install --upgrade transformers torch sentence-transformers

Or manually clear cache and retry:

# Windows
rmdir /s /q %USERPROFILE%\.cache\huggingface

# macOS/Linux
rm -rf ~/.cache/huggingface

Issue: `Out of Memory` when loading Flan-T5

Solution:

• —Ensure you have 16GB+ RAM
• —Close other applications
• —If still failing, the model will automatically fall back to GPT-4 only

Issue: Streamlit won't start

Solution:

# Check if Streamlit is installed
streamlit --version

# Reinstall Streamlit
pip uninstall streamlit
pip install streamlit==1.31.0

# Try running from project directory
cd ResearchForge
streamlit run app.py

📁 Project Structure

ResearchForge/
├── app.py                      # Streamlit web interface
├── multi_agent_system.py       # Core logic (1932 lines)
├── requirements.txt            # Python dependencies
├── .env                        # API keys (create this)
├── .gitignore                  # Git ignore rules
├── README.md                   # Project documentation
├── PROJECT_DOCUMENTATION.md    # Detailed technical docs
└── images/                     # Architecture diagrams

🤝 Support

Issues? Open an issue at: https://github.com/Radia-987/ResearchForge/issues

Questions? Check the detailed documentation: PROJECT_DOCUMENTATION.md

🎯 Next Steps

After successful installation:

1. 1.✅ Test with a simple query (e.g., "machine learning")
2. 2.✅ Review the generated literature review
3. 3.✅ Examine the identified research gaps
4. 4.✅ Explore the generated hypotheses
5. 5.✅ Check experiment design suggestions

📝 License

This project is open source and available for educational and research purposes.

🙏 Acknowledgments

AI Models Used:

• —OpenAI GPT-4 & GPT-3.5-turbo
• —Google Flan-T5-Large
• —Sentence-BERT (all-MiniLM-L6-v2)

Data Sources:

• —arXiv API
• —Google Custom Search
• —ClinicalTrials.gov

🎉 You're ready to start! Run `streamlit run app.py` and explore AI-powered research analysis!