The HVSA is a system designed for thesis purposes of EARIST MANILA students. The primary contributors to this project include software developer Jimuel Flojera and team members Jenry Kim, Roger Rosal Jr., and Lovely Mae Pango. The system utilizes Flask as the backend server and incorporates a fine-tuned model of RoBERTa for text classification. Its main objective is to assess the authenticity of academic works submitted by students, distinguishing between works created by humans and those generated by artificial intelligence.
- Text Classification: The system utilizes a fine-tuned RoBERTa model to perform text classification on academic works.
- Flask Backend: The backend server is implemented using Flask, providing a robust and scalable foundation for the system.
- User Interface (UI): A simple and intuitive UI is provided for users to submit academic works and receive classification results.
- Python 3.6 or higher
- Flask
- Transformers library
- Other dependencies (specified in requirements.txt)
-
Clone the repository:
git clone https://github.com/Jimuell12/Flask-Text-Detection.git
-
Install dependencies:
pip install -r requirements.txt
-
Run the Flask application:
python app.py
The application will be accessible at
http://localhost:8080.
- Access the system through a web browser.
- Upload an academic work (text document) for analysis.
- Receive the classification result indicating whether the work is likely created by a human or AI.
- Special thanks to the Hugging Face community for providing pre-trained models and resources.
- We extend our heartfelt gratitude to our thesis adviser, Joevy Fajardo, for invaluable guidance, support, and mentorship throughout the research process.
- We also extend our gratitude to Professor Merlita Latip, our thesis course professor, for her insightful feedback, encouragement, and dedication to our academic growth.
For inquiries, issues, or collaborations, please contact Jimuel Flojera at [email protected].