COMP90051-P2

This repository contains the codebase for Project 2 of the subject COMP90051 – Statistical Machine Learning, Semester 1, 2025, at the University of Melbourne. This project is part of a group submission for Group 28.

Team Members:

• Ankita Holey
• Richardo Husni
• Sarathi Thirumalai Soundararajan

The objective of this project is to build a classification model to distinguish between machine-generated and human-written text, leveraging domain adaptation and imbalance handling techniques. The project was conducted as part of a Kaggle competition outlined in the Project Specification PDF, which includes additional details on data format, performance evaluation, and submission criteria.

---

Setup Instructions

1. Clone the Repository

If you haven't already cloned the repository, run:

git clone https://github.com/sthirumalais/COMP90051-A2.git
cd COMP90051-A2

2. Create the Conda Environment

Ensure you have Conda (Anaconda or Miniconda) installed, then create the environment:

conda env create -f environment.yml

This will create a Conda environment named p2 with all required dependencies.

3. Activate the Environment

Activate the newly created environment:

conda activate p2

4. Install the Jupyter Kernel

Register the Jupyter kernel:

python -m ipykernel install \
--user \
--name COMP90051-P2 \
--display-name "COMP90051-P2"

5. Launch JupyterLab or Notebook

Verify that the kernel COMP90051-P2 is available:

jupyter lab
# or
jupyter notebook

Select the COMP90051-P2 kernel when prompted.

---

Notebook Execution Order

To ensure reproducibility and correctness, run the notebooks in the following order:

1. Pre-Processing.ipynb
2. Feature-Engineering.ipynb
3. Then, run any of the model notebooks below as needed:

• Model-LR.ipynb
• Model-DANN.ipynb
• Model-MTL-1.ipynb
• Model-MTL-2.ipynb

---

Notes

• Ensure your kernel is set to COMP90051-P2 in each notebook.
• If you encounter issues with dependencies, try updating Conda and re-creating the environment.

---

Happy coding!