AI Text Detection

AI-Generated Text Detection

This project is designed to classify text as either AI-generated or human-written using deep learning techniques. It combines text embeddings with numerical linguistic features to improve accuracy.

Linguistic Feature Extraction

Linguistic Feature Extraction

Architecture

Linguistic Feature Extraction

Features

  • Uses Sentence Transformers for text embeddings
  • Incorporates 27 linguistic features for enhanced classification
  • Built with PyTorch and Streamlit for an interactive user interface
  • Real-time text analysis with probability scores

Technologies Used

  • Python, PyTorch, NLP models
  • Streamlit UI for real-time interaction
  • Scikit-learn for feature engineering

Demo

Screenshots

AI Text Detection UI