History and Culture Machine Learning (ML) for Tracking Fashion Trends: Documenting the Frequency of the Baseball Cap on Social Media and the Runway Rachel Rose Getman 1 , Denise Nicole Green 1 , Kavita Bala 2 , Utkarsh Mall 2 , Nehal Rawat 2 , Sonia Appasamy 2 , and Bharath Hariharan 2 Abstract With the proliferation of digital photographs and the increasing digitization of historical imagery, fashion studies scholars must consider new methods for interpreting large data sets. Computational methods to analyze visual forms of big data have been underway in the field of computer science through computer vision, where computers are trained to “read” images through a process called machine learning. In this study, fashion historians and computer scientists collaborated to explore the practical potential of this emergent method by examining a trend related to one particular fashion item—the baseball cap—across two big data sets—the Vogue Runway database (2000– 2018) and the Matzen et al. Streetstyle-27K data set (2013–2016). We illustrate one implementation of high-level concept recognition to map a fashion trend. Tracking trend frequency helps visualize larger patterns and cultural shifts while creating sociohistorical records of aesthetics, which benefits fashion scholars and industry alike. Keywords machine learning, trend analysis, Instagram, Vogue Runway, baseball cap, fashion trend, interdisci- plinary collaboration, computer vision In the digital age, access to millions of images, media assets, and other information (e.g., location, browsing history, and other identity markers) could provide rich data for fashion trend identification and analysis. In this study, researchers explored the potential of computer vision—that is, the 1 Department of Fiber Science & Apparel Design, College of Human Ecology, Cornell University, Ithaca, NY, USA 2 Department of Computer Science, Cornell University, Ithaca, NY, USA Corresponding Author: Rachel Rose Getman, Department of Fiber Science & Apparel Design, College of Human Ecology, Human Ecology Building, Cornell University, Ithaca, NY 14853, USA. Email: rrg84@cornell.edu Clothing and Textiles Research Journal 1-16 ª 2020 ITAA Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/0887302X20931195 journals.sagepub.com/home/ctr