Eye for Feature Engineering

This problem was initially used in MAIO 2025, and has been slightly modified for distribution.

This problem was originally designed to be solved within 2hrs, and was graded completely by hand. Use of all LLM assistance including reasoning models were allowed.

Preview

You are given a peculiar dataset passed through a logistic regression classifier. Notice that the baseline example provided above scores 0 for precision, recall and F1 score.

Do what you can to raise the F1 score as much as possible through feature engineering!

Download

Right click this link and select “save as”: maio_2025_eye_for_feature_engineering_public.ipynb

Editorial

To be updated!

Citation

You are welcome to publish your own solutions to this problem! Please use the provided citation below:

@misc{maio2025_eye_for_feature_engineering,
    title        = {Eye For Feature Engineering},
    author       = {Tan, Nian Wei},
    year         = {2025},
    howpublished = {Blog},
    url          = {https://aiolympiad.my/past-problems/maio2025_eye_for_feature_engineering},
}