🔍 AI-Based Research Assistant

An AI-powered research assistant that automates literature reviews, summarizes papers, analyzes PDFs, and conducts web-based research efficiently.

---

🚀 Project Overview

This AI-powered application leverages cutting-edge technologies to assist researchers by:

• 📚 Searching Research Papers: Fetches research papers from Google Scholar based on a given topic.
• 📝 Generating Literature Reviews: Summarizes key findings, methodologies, and implications from the retrieved papers.
• 📄 Analyzing PDF Research Papers: Extracts text from uploaded PDFs, uses RAG (Retrieval-Augmented Generation), and answers user questions about the content.
• 🌐 Web-Based Research: Conducts research on any topic using DuckDuckGo and Wikipedia APIs, providing a structured summary.

---

🧠 Key Functionalities

1. Search for Research Papers
   • Retrieves research papers using Google Scholar API (scholarly).
   • Generates structured literature reviews and comprehensive summaries.
2. Analyze PDF Research Papers
   • Extracts and processes PDF content using PyMuPDF.
   • Sets up a RAG system with FAISS to answer questions about the uploaded paper.
3. Web Research
   • Uses DuckDuckGo and Wikipedia APIs for topic-based web research.
   • Summarizes the results in a structured format using LangChain.

---

🛠️ Tech Stack

• Backend: Python, LangChain, FAISS, PyMuPDF
• Frontend: Streamlit
• AI Models: Claude-3 Sonnet (via Anthropic API), SentenceTransformer (all-MiniLM-L6-v2)
• APIs & Tools:
  • Google Scholar API (scholarly)
  • DuckDuckGo API
  • Wikipedia API
• Libraries: LangChain, FAISS, anthropic, scholarly, pyttsx3, PyMuPDF, re, numpy, pandas

---

📄 Installation & Setup

1. Clone the Repository

git clone https://github.com/atharvad38/AI_Based_Research_Assistant.git
cd AI_Based_Research_Assistant

2. Create a Virtual Environment

# Create a virtual environment
python3 -m venv venv

# Activate virtual environment
# Linux/Mac
source venv/bin/activate
# Windows
venv\Scripts\activate

3. Install Required Dependencies

pip install -r requirements.txt

4. Set API Keys

Create a .env file in the root directory and add:

ANTHROPIC_API_KEY=your-anthropic-api-key

---

▶️ Usage

1. Run the Application

streamlit run main.py

2. Explore Features

• Search for Research Papers: Enter a research topic and retrieve papers.
• Analyze PDF Paper: Upload a research paper PDF and ask questions.
• Web Research: Search the web for any topic and get a structured summary.

---

📚 How It Works

📝 PDF Analysis with RAG

• Chunks text from PDFs using PyMuPDF and creates vector embeddings with SentenceTransformer.
• FAISS indexes these embeddings to enable semantic search.
• Retrieves relevant chunks and uses Claude AI to generate answers.

🔎 Web Research with LangChain

• Uses DuckDuckGo and Wikipedia APIs.
• Claude AI processes retrieved content and formats it as a structured summary.

⚠️ Disclaimer

This tool is intended for academic and research purposes. Always cross-check the information from multiple sources for critical research.

---

🔗 GitHub Repository: AI-Based Research Assistant