This repository is set up to include reproducible code and data for the paper Expertise Prediction of Tetris Players Using Eye Tracking Information. The repository contains 5 Python files: split.py, train.py, test.py, attribute.py and plot.py.

split.py provides the code used to generate the files in the data/ts folder. It cannot be run for now, as we omitted the original participant data.

train.py runs the classification models using 5-fold cross validation. Most methods were imported from the aeon package, but the MR-PETSC was adapted from the repository https://bitbucket.org/len_feremans/petsc/src/master/. The results for each method are stored in a csv file in the results folder.

test.py loads the results from the results folder for each method and provides several metrics, it also contains the code for reproducing the confusion matrices in the paper.

attribute.py creates the figure with attribution.

Lastly, plot.py contains the code for the other figure in the paper.