This year, the conference was held in San Francisco. And for the first time, the program included a workshop on the fashion industry, titled: “Machine Learning meets Fashion”. Specialists from academia, industry and startups working at the intersection of fashion and data mining and knowledge discovery gathered to share their perspectives and insights.
As the program notes pointed out: “Fashion is a multi-billion dollar industry with social and economic implications worldwide. The fashion industry has traditionally placed high value on human creativity and has been slower to realize the potential of data analytics. With the advent of modern cognitive computing technologies….and vast amounts of (structured and unstructured) fashion data the impact on the fashion industry could be transformational.”
In a joint presentation by Oh and Eberhard, they presented a talk about: “Challenges of quantifying fashion data: Creativity, art and emotions.’’ The goal was to bridge the gap between data analysts, who speak largely in their own specialized vocabulary and the fashion community, which has its own creative processes of communication.
Topics covered included the challenges of quantifying fashion data and the problems of how to apply big data analysis methods to the fashion industry.
The groups working in this field are trying to elaborate a common vocabulary for fashion companies, create algorithms for recommendations on e-commerce websites and do data mining from social media platforms for trend forecasting.
Fashion, as a field at the border of art and industry, is a new and exciting domain for data scientists to apply their methods. Real fashion world, designers and creative consumers (street fashion) provide eclectic ever-changing content that science and technology are trying to optimize in order to increase sales and decrease the waste of over-production.
Areas of overlap in the new interest in how best to encompass both disciplines include the following: Forecasting fashion trends, influencer analytics, visual search, natural language processing and style recommendation algorithms.
For data scientists it is necessary to understand the natural life-cycle of a fashion garment before applying science in order to accelerate or alter it.
Recent projects by companies including Microsoft, Amazon, Zalando and Stitch Fix, are examples of successful collaborations between giants of technology and academics exploring the potential of quantifying fashion data.
Big data algorithms have become essential features of many retail websites, where they help to resolve the challenges of visual search. It has also entered the field of fashion journalism through the social media applications and influencer analytics.
So it comes as no surprise that the School of Fashion at Academy has also followed this important discourse and upgraded its courses to now include data-heavy topics such as Data Journalism, Social Listening, and Social Media Data Analytics.
Other topics and speakers at the conference included “Making fashion recommendations with human-in-the-loop machine learning,’’ by Brad Klingenberg, Director of Data Science at Stitch Fix; “Recommendation and Opinion Making with Visual Signals’’ by Julian McAuley, Assistant Professor, Computer Science at UC San Diego and presentations on “Decoding Fashion Contexts Using Word Embeddings’’ and “Detection of fashion trends and seasonal cycles through the analysis of implicit and explicit client feedback.”
The two seemingly disparate worlds of fashion and technology are growing closer and closer together. And the KDD conference was an important step of uniting them through the common goals of expanding Knowledge, Discovery and Data Mining.
Text by Elena Eberhard, Public Relations & Special Events Manager at the School of Fashion, Academy of Art University.