Textile and Apparel Industries

Misconceptions and Satisfaction in Female Body Shapes: Utilizing Supervised Machine Learning Algorithms for Identifying Body Shapes

Authors: Uikyung Jung (University of Central Oklahoma) , Chanmi Hwang (NC State University) , Minyoung Suh (North Carolina State University)

Abstract

This study analyzed the congruences between objective and perceived female body shapes and further examined the influence of body shape on body shape congruence and body satisfaction. A predictive model was developed to determine objective body shapes, employing multinomial logistic regression (MLR), random forest (RF), and support vector machine (SVM) techniques on the SizeUSA dataset. With the MLR model demonstrating superior accuracy compared to other algorithms, it was chosen to categorize the objective body shape within a new dataset obtained from 212 female participants. Subsequently, the objective classifications of body shapes were juxtaposed with the subjective self-identifications of body shapes by the participants.

Keywords: Body Shape, Body Satisfaction, Garment Fit, Machine Learning, Female Figure Identification Technique

How to Cite: Jung, U. , Hwang, C. & Suh, M. (2024) “Misconceptions and Satisfaction in Female Body Shapes: Utilizing Supervised Machine Learning Algorithms for Identifying Body Shapes”, International Textile and Apparel Association Annual Conference Proceedings. 80(1). doi: https://doi.org/10.31274/itaa.17584