Using Conceptual Modeling to Support Machine Learning

摘要

With the transformation of our society into a "digital world," machine learning has emerged as an essential approach to extracting useful information from large collections of data. However, challenges remain for using machine learning effectively. We propose that some of these can be overcome using conceptual modeling. We examine a popular cross-industry standard process for data mining, commonly known as CRISP-DM Directions, and show the potential usefulness of conceptual modeling at each stage of this process. The results are illustrated through an application to a management system for drug monitoring. Doing so demonstrates that conceptual modeling can advance machine learning by: (1) supporting the application of machine learning within organizations; (2) improving the usability of machine learning as decision tools; and (3) optimizing the performance of machine learning algorithms. Based on the CRISP-DM framework, we propose six research directions that should be explored to understand how conceptual modeling can support and extend machine learning.