Datenpaket: Experimental Data for the Paper "Quantifying Domain-Application Knowledge Mismatch in Ontology-Guided Machine Learning"

These are experimental data for the paper: Pawel Bielski, Lena Witterauf, Sönke Jendral, Ralf Mikut, Jakob Bach. 2024. Quantifying Domain-Application Knowledge Mismatch in Ontology-Guided Machine Learning, 16th International Conference on Knowledge Engineering and Ontology Development (KEOD), Porto, Portugal. The data consist of the results from the experiments and the corresponding code to create plots used in the paper.

Folder Structure

• Use Case 1 Computer Vision: contains jupyter notebook files to compute refinement scores for the three image classification dataset, the pickle files with the computed refinement scores, and the file to generate plots used in the paper.
• Use Case 2 Healthcare: contains two subfolders
• experiments_code contains the code used to reproduce the experiments on the MIMIC-III Dataset.
• experimental_results_code contains the jupyter notebook files to compute the refinement scores for the next sequence prediction and risk prediction tasks, the pickle files with the computed refinement scores, and the file to generate plots used in the paper.