By Drew Hendrickson, Tilburg School of Humanities and Digital Sciences at Tilburg University
In this talk I’ll present a gentle introduction to the concepts of data science and machine learning with a focus on applications to experience sampling data and the insights these analyses can provide. I’ll highlight the differences between inferential statistics, network estimation, and machine learning analyses as well as discuss how each of these methods (usually) answer qualitatively different but related research questions. Finally, I will outline the various stages of a standard data science analysis pipeline for experience sampling data from a recent project on predicting adolescent stress (1). This discussion will emphasize the challenges ESM data can cause for machine learning analyses as well as the potential impact and utility of these prediction systems.