An Augmented Method of Selecting Fashion Talent by Adding Social Media Characteristic

Abstract

Fashion model plays an important role in presenting designer's work by showing how the cut of fabric interplayed with the body. In selecting the fashion models, the agency considers physical characteristics that express the aesthetic. Another subjective advantage of fashion model is the appearance on professional network. However, the emerging of social media culture has revolutionized fashion industry in producing and consuming fashion. In term of fashion modelling, social media open the opportunities for talent to "breaking in" and "getting discovered". Prior research has dealt with predicting success based on social media presence. Thus, in this paper we construct additional social media activity to predict a fashion model success. We examine prediction using classification task by utilizing Random Forest and Support Vector Machine. Our research finds that social media activity improves the accuracy by 4.55% increasing up to 84.55% performed by Random Forest.