🧠 AI Text Detector & Analysis System

---

📌 Overview

An advanced AI-powered text analysis system that detects whether content is AI-generated or human-written using a hybrid architecture combining:

• 🤖 Transformer-based deep learning
• ✍️ Stylometric feature analysis
• 📊 Machine learning classification

It also includes:

• Grammar checking
• Style analysis
• Plagiarism detection

---

🎯 Key Features

✨ AI vs Human Detection ✨ Hybrid Model (DeBERTa + Stylometry) ✨ Grammar & Writing Feedback ✨ Plagiarism Detection ✨ Long Document Support (Chunking) ✨ Chrome Extension Integration ✨ GPU Acceleration (CUDA)

---

🏗️ System Architecture

🔄 Pipeline

🧩 Core Modules

🧠 AI Detection Engine

• Uses DeBERTa Transformer
• Extracts contextual embeddings
• Works as semantic understanding layer

---

✍️ Stylometric Analysis

• Captures writing style patterns:

  • Sentence length
  • Word complexity
  • Vocabulary richness
  • Punctuation usage

---

📊 Grammar Analysis

• Powered by LanguageTool
• Detects grammatical errors
• Suggests corrections

---

📚 Plagiarism Detection

• Integrated with external API
• Detects similarity with online content
• Returns plagiarism score

---

⚡ ML Infrastructure

• Hybrid model:

  • Deep Learning + Traditional ML

• Supports:

  • GPU (CUDA)
  • Large text processing

---

🛠️ Tech Stack

🔹 Backend

• FastAPI
• PyTorch
• Hugging Face Transformers
• scikit-learn
• spaCy
• NumPy

---

🔹 Frontend

• Next.js
• React

---

🔹 Other

• CUDA (GPU acceleration)
• LanguageTool (Grammar checking)
• External API (Plagiarism)

---

🔮 Future Scope

• Multilingual support
• Explainable AI (highlight AI text)
• Real-time editor integration
• Improved long-text understanding
• Continuous model updates

🚀 Steps to Run the Project (Local)

1️⃣ Clone the repository

git clone <repository-url>

---

🧠 Run Backend (FastAPI)

2️⃣ Go to backend directory

cd backend

---

3️⃣ Create & activate virtual environment

Windows

python -m venv venv
venv\Scripts\activate

macOS / Linux

python3 -m venv venv
source venv/bin/activate

---

4️⃣ Install backend dependencies

pip install -r requirements.txt

---

5️⃣ Start backend server

uvicorn app:app --reload

Backend runs at:

http://127.0.0.1:8000

---

🌐 Run Frontend (Next.js)

6️⃣ Open new terminal & go to project root

cd ai-text-detector

---

7️⃣ Install frontend dependencies

npm install

---

8️⃣ Start frontend server

npm run dev

Frontend runs at:

http://localhost:3000

---

🔄 Run Summary

Service	Command	Port
Backend	uvicorn app:app --reload	8000
Frontend	npm run dev	3000

---

🧩 Run Chrome Extension (Local)

9️⃣ Go to extensions directory

From the project root:

cd extensions

Ensure this folder contains a manifest.json file.

🔟 Open Chrome Extensions page

Open Google Chrome and navigate to:

chrome://extensions

1️⃣1️⃣ Enable Developer Mode

Turn Developer mode ON (top-right corner).

1️⃣2️⃣ Load the extension

Click Load unpacked

Select the extensions folder

Click Open

✅ The extension will be successfully loaded into Chrome.

1️⃣3️⃣ Use the extension

Click the 🧩 Extensions icon in the Chrome toolbar

Pin your extension

Open any webpage and use the extension features

🔄 Complete Run Summary

Service	Command / Action	Port
Backend	uvicorn app:app --reload	8000
Frontend	npm run dev	3000
Chrome Extension	Load Unpacked (extensions/)	—