AI-Admissions-Detector

Code for application and training algorithms described in:

Yijun Zhao, Fernando Martinez, Haoran Xue, Gary M. Weiss (2024) "Admissions in the Age of AI: Detecting AI-Generated Application Materials in Higher Education"

Repository Structure

This repository is organized as follows:

src Directory

The src folder contains scripts used for the generation of prompts and the training and analysis of AI models.

• LORPromptsMaker.py: Generates prompts for letters of recommendation.
• SOIPromptsMaker.py: Generates prompts for statements of intent.
• TrainingAndAnalysis.py: Handles the training and analysis of models.

app Directory

The app directory encompasses all the necessary components to run the application.

• app.py: Main application entry point.
• custom_models.py: Contains custom transformer-based models.
• requirements.txt: Lists all dependencies required to run the application.
• Dockerfile: Dockerfile for building the application container.
• .streamlit: Contains Streamlit configuration files (if applicable).

models Subdirectory

The models subdirectory within app contains baseline models for machine learning operations.

• baseline_model_lr.joblib: Baseline logistic regression model.
• baseline_model_lr2.joblib: Second logistic regression baseline model.
• baseline_model_nb.joblib: Baseline Naive Bayes model.
• baseline_model_nb2.joblib: Second Naive Bayes baseline model.