By combining social media and traditional algorithms with machine learning, “fashion discovery” search engine Fynd offers a virtual “personal shopper to everyone,” the company said as it publicly launches its Web site, Fynd.me, today.
“Fynd allows you to locate the dresses you’re looking for without the confusion of filters,” the company said this morning. “Fynd is a fashion site where users find products they are seeking by simply entering a description of what they are looking for, using everyday English or ‘liking’ suggested items.” The criteria for the “likes” include color, cut and occasion, among other descriptions.
The site’s engine then goes into robot mode to “curate and suggest” various styles based on what was liked, and then it “instantly connects shoppers with trusted luxury retailers” such as Nordstrom Inc. and Bloomingdale’s. The more “likes” that are inputted into the site results in a deeper, richer set of recommendations.
Fynd cofounder Charese Embree said the “traditional online shopping experience takes all the joy out of finding the right look – it’s become a real chore.”
Embree said Fynd was designed to be a “concierge service that reconnects consumers with that feeling they get from the visual experience of shopping with a stylist in person.” Embree added that “intuitive social elements” and functionality were done to meet the demands of shoppers.
The initial launch will be in women’s dresses, and then will expand to other product segments as well as “styles for men, women and children by late 2016,” the company, which is based in New York, noted.
Machine learning began as a subset of computer science applications, but is seeing a resurgence, which “is due to the same factors that have made data mining and Bayesian analysis more popular than ever,” said researchers at SAS in a recent report, adding that those factors include a larger volume and variety of data as well as more affordable data storage.