Sign In

Communications of the ACM

ACM TechNews

Retailers Use AI to Improve Online Recommendations for Shoppers

View as: Print Mobile App Share:
A view of e-commerce.

Online retailers like Wayfair, Etsy, and Pinterest are ratcheting up efforts to leverage data from a surge in e-commerce to supercharge their search and recommendation engines.

Credit: Harry Campbell

Online retailers have taken steps during the pandemic to improve their search and recommendation engines by building predictive models with high levels of accuracy.

Gartner's Bob Hetu said online retailers increasingly are turning to computer vision models to index products in their virtual catalog using visual cues and natural-language processing to aggregate and learn from words used in shoppers' searches.

These tools rely on algorithms powered by machine learning, which continually use the results they produce to fine-tune the models.

Wayfair's computer-vision algorithm captures design features, materials, styles, color, vintage, and other elements to tag items, and the richer the list of keywords tagged to a product, the more relevant the search results.

Meanwhile, Etsy's search engine uses natural-language process to learn from customers' past queries and purchases, and the company plans to train algorithms to combine product and search data over time to gauge a shopper's overall taste.

From The Wall Street Journal
View Full Article - May Require Paid Subscription


Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account