An Effective Research Method to Predict Human Body Type Using an Artificial Neural Network and a Discriminant Analysis

Artificial intelligence (AI) technology can be an effective solution to decision-making in the apparel field; however, few studies have used the AI-based methodology for categorizing and predicting human body types. The researchers in this study demonstrate the accuracy and effectiveness of the artificial neural network (ANN) as an approach to AI technology to predict women's body types by comparing the predictive accuracy rate between a statistical discriminant analysis and an ANN. It was observed that the ANN (94.7 %) can predict body types more accurately than the discriminant analysis (83.5 %). The ANN is a more effective method than a statistical analysis because the ANN requires fewer input measurement variables than the discriminant analysis to predict body types. Comparing the predictive accuracy rate based on a different number of learning times and input variables, the researchers offer a new methodological guideline to use in future research studies in this field.