🧠 Vertical.ai Notebook

###Vertical.ai for Papers, Notes & Interactive Mind Maps

Vertical.ai Notebook is a document-centric AI research workspace inspired by Google Notebook-LM.
It enables researchers to search, import, and reason over academic papers using citation-grounded AI, a dynamic interactive mind map, and a clean notebook-style interface.

Built with a strong focus on trust, explainability, and real research workflows — not just a chatbot.

---

✨ Key Highlights

• 📚 Notebook-LM style workflow
• 🔍 Free academic paper discovery (OpenAlex)
• 💬 Citation-aware research chat (RAG)
• 🧠 Dynamic, expandable mind maps
• 🧩 Multiple structured views from the same source
• 🔐 Secure API handling (no hard-coded secrets)

---

🚀 Core Features

🔍 Academic Source Search & Import

• Topic-based academic search using OpenAlex (100% free, no API key)
• Citation-based ranking (highly cited papers first)
• Fast research previews (title, abstract, citation count)
• One-click selective import into the notebook

---

💬 Citation-Aware Research Chat

• Ask questions only on imported sources
• Retrieval-Augmented Generation (FAISS + embeddings)
• Answers include explicit citations
• Persistent notebook-style conversation (not a chatbot reset)

---

🧠 Dynamic Interactive Mind Map

• Notebook-LM–style full-screen modal
• Workflow-style rectangular nodes
• Lazy loading (expand nodes on click)
• Click any node → detailed explanation appears in chat
• Zoom, pan, focus & breadcrumb navigation

---

🧩 Source Guide Cards

Multiple representations over the same knowledge base:

• 🧠 Mind Map
• 📇 Flashcards
• 📽️ Slide Deck outline
• 📊 Data Tables

One source of truth → many cognitive views.

---

🏗️ System Architecture

Frontend (React + Vite) ↓ Backend (FastAPI) ↓ Search Layer (OpenAlex + Tavily) ↓ Ingestion Pipeline (PDF → Chunks → Embeddings) ↓ Vector Store (FAISS) ↓ LLM Reasoning (Groq) ↓ Knowledge Graph (Neo4j)

---

🧪 Tech Stack

Frontend

• React + Vite
• Notebook-style 3-panel layout
  (Sources | Chat | Source Guide)

Backend

• FastAPI
• Modular REST APIs

AI / ML

• Groq LLM — reasoning & summarization
• HuggingFace embeddings (SciBERT / E5 – local)
• FAISS — vector similarity search

Search & Discovery

• OpenAlex — academic papers (free)
• Tavily — optional web search

Knowledge Representation

• Neo4j (local knowledge graph)

---

🔐 Security & API Management

• ❌ No API keys hard-coded in source code
• ✅ Environment-based configuration
• ✅ .env file ignored via .gitignore

Environment Variables

GROQ_API_KEY=your_groq_api_key
TAVILY_API_KEY=your_tavily_api_key


Installation & Setup
1️⃣ Clone the Repository
git clone https://github.com/yourusername/ResearchAI-Notebook.git
cd ResearchAI-Notebook

2️⃣ Backend Setup
cd backend
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -r requirements.txt
uvicorn main:app --reload