README.md

metadata

title: AI vs Human Text Classifier
emoji: 🚀
colorFrom: indigo
colorTo: pink
sdk: streamlit
sdk_version: 1.32.2
app_file: app.py
pinned: false

🧠 AI vs Human Text Detector

This project is a machine learning-based system designed to distinguish between text written by a human and that generated by an AI language model (e.g., ChatGPT). It uses deep learning and text embeddings to analyze writing patterns and classify them accurately.

---

📌 Features

• ✅ Binary classification: AI-generated vs Human-written text
• 📈 Achieved 92% accuracy and 0.89 F1-Score
• 🔍 Embedding + Deep Learning model
• 📊 Evaluation on real-world prompts and datasets
• 🧪 Trained and tested using clean, balanced samples

---

🚀 Live Demo

🔗 Try it on Hugging Face Spaces

Enter a piece of text and the model will predict whether it's written by an AI or a human.

---

📂 Repository Structure

.
├── train_model.ipynb      # Notebook to preprocess and train the model
├── requirements.txt       # (Optional) Dependencies
└── README.md              # Project overview and instructions