Skip to content

This repository contains the code and scripts used in my thesis project, which aims to build various BERT and LSTM models to detect AI generated texts in both English and Arabic languages.

Notifications You must be signed in to change notification settings

haashhish/AI-Detection-Tool

Repository files navigation

Thesis Project: AI Plagiarism Detection Tool

The repository includes the following files:

  • app.py: Main application script for deploying the models.
  • bert_mini_arabic.py: Script for training and evaluating a BERT model on Arabic text data.
  • bert.py: Script for training and evaluating a BERT model on English text data.
  • FINAL_model.py: Final script for training the model, incorporating optimizations and adjustments based on earlier experiments.
  • FINAL_testing.py: Script for testing the final trained model on a separate test dataset.
  • lstmArabic.py: Script for training and evaluating an LSTM model on Arabic text data.
  • testArabic.py: Script for testing the LSTM model on Arabic text data.
  • lstm.py: Script for training and evaluating an LSTM model on English text data.
  • testEnglish.py: Script for testing the LSTM model on English text data.

Prerequisites

  • Python 3.7 or higher

Weights:

Datasets:

  • Datasets used in this project are not included in the repository and should be requested via email. Please send your request to mohamedh_4@aucegypt.edu with a brief explanation of your use case.

Note: This project is my thesis work at AUC during Spring 2024 and is intended for educational and research purposes. Any use of the provided code should appropriately cite this repository and the associated thesis. If you have any questions or requests, feel free to contact me at mohamedh_4@aucegypt.edu

About

This repository contains the code and scripts used in my thesis project, which aims to build various BERT and LSTM models to detect AI generated texts in both English and Arabic languages.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages