Jadwal Sholat

Memuat jadwal sholatโ€ฆ

Pendidikan editorial

Open AccessOA2022

Applications of Machine Learning to Behavioral Sciences: Focus on Categorical Data

This perspective article examines the limited use of machine learning on non-imaging, categorical behavioral data and advocates for its broader application to complement traditional inference-based models.
Pegah Dehghan; Hany Alashwal; Ahmed A. Moustafaยท Discover Psychologyยท 2022ยท DOI 10.1007/s44202-022-00027-5

The core problem

In the last two decades, advancements in artificial intelligence and data science have attracted researchers' attention to machine learning. Growing interests in applying machine learning algorithms can be observed in different scientific areas, including behavioral sciences. However, most of the research conducted in this area applied machine learning algorithms to imaging and physiological data such as EEG and fMRI, and there are relatively limited non-imaging and non-physiological behavioral studies which have used machine learning to analyze their data. Therefore, in this perspective article, the authors aim to (1) provide a general understanding of models built for inference, models built for prediction (i.e., machine learning), methods used in these models, and their strengths and limitations; (2) investigate the applications of machine learning to categorical data in behavioral sciences; and (3) highlight the usefulness of applying machine learning algorithms to non-imaging and non-physiological data (e.g., clinical and categorical) data and provide evidence to encourage researchers to conduct further machine learning studies in behavioral and clinical sciences.

Innovation

As a perspective article, the methodology involves a conceptual review and synthesis of the literature rather than empirical data collection. The authors compare and contrast inferential models and predictive (machine learning) models, discussing their respective methods, strengths, and limitations. They then survey applications of machine learning to categorical data in behavioral sciences, with a focus on non-imaging and non-physiological data. The article does not present new experimental data; instead, it builds an argument based on existing research to encourage further machine learning studies in behavioral and clinical sciences.
Introduction
In the last two decades, advancements in artificial intelligence and data science have attracted researchers' attention to machine learning. Growing interests in applying machine learning algorithms can be observed in different scientific areas, including behavioral sciences. However, most of the research conducted in this area applied machine learning algorithms to imaging and physiological data such as EEG and fMRI, and there are relatively limited non-imaging and non-physiological behavioral studies which have used machine learning to analyze their data. Therefore, in this perspective article, the authors aim to (1) provide a general understanding of models built for inference, models built for prediction (i.e., machine learning), methods used in these models, and their strengths and limitations; (2) investigate the applications of machine learning to categorical data in behavioral sciences; and (3) highlight the usefulness of applying machine learning algorithms to non-imaging and non-physiological data (e.g., clinical and categorical) data and provide evidence to encourage researchers to conduct further machine learning studies in behavioral and clinical sciences.

Why it matters

The authors discuss the strengths and limitations of inference-based models versus prediction-based (machine learning) models. Inference models focus on explaining relationships and testing hypotheses, while machine learning models prioritize predictive accuracy and can handle complex, high-dimensional data. The article argues that machine learning can complement traditional methods in behavioral sciences, especially for categorical data, by uncovering patterns and improving prediction. However, challenges such as interpretability, overfitting, and data quality are noted. The authors conclude by encouraging researchers to apply machine learning to non-imaging and non-physiological data to advance behavioral and clinical research.

Who should read this

Pembaca Research Digest dan praktisi rumpun ini.

Opening member contentโ€ฆ