Kavita Bala and her team at Cornell University used artificial intelligence to scour social media for fashion trends and insights. Here are a few of the StreetStyle project’s findings:

• White and black clothing reveals a high periodic trend. White reaches its peak in September, while black is much more common during the winter, particularly in January. Cities in the southern hemisphere are on the flip side of this pattern, suggesting a correlation between season and color of clothing.
• Darker colors such as brown are popular in winter, while blue is popular in the summer.
• Turns out, many people follow mom’s advice about not wearing white after Labor Day. The project showed a significant decrease in white clothing beginning in mid-September.

StreetStyle is a project that uses AI to mine social media posts and identify fashion trends geographically.

• Red has several spikes each year, near the end of October for Halloween and in December around Christmas. There’s also a huge increase in red clothing in China during Chinese New Year.
• Hats are more popular in colder places. In Oman, hats – particularly the kuma and massar – are prevalent since it’s an essential element of the country’s male national clothing.
• Countries farther north tend to wear more jackets. In South America, more jackets are found further west around Bolivia and Colombia.
• Blue collared shirts, plaid shirts and black T-shirts are common around the world throughout the year.

Many of the insights gleaned from the StreetStyle project seem obvious. However, it’s a great example of how AI can easily digest readily available chunks of data. Though the initial study had several limitations – leaving out countries with no internet access, for example – Bala and her team plan to expand and improve upon their style study in the future.

Product Hub
Find the latest in quality products, must-know trends and fresh ideas for upcoming end-buyer campaigns.