Analysis of Medical Data and Machine-Learning Algorithms from the Perspective of Public-Goods Models of Data-Provision Decision Making

Abstract

Listen

Translate article

With the development of the machine-learning field, the early detection of diseases and development of new drugs are being conducted by learning vast amounts of medical data. The authors theoretically analyze medical-data holders’ provision decision from the perspective of a public-goods model. Through a meta-analysis of medical datasets on paperswithcode.com, the authors explore the relationship between medical-data sample size and machine-learning models’ scores using such data. The result suggests that an increase in the number of data providers leads to higher model performance, depending on the score measurement selected.