Yelp's Popular Dishes frees you from the agony of choosing what to eat

If Yelp has its say, you'll never have to ask a waiter "What's good here?" ever again.

The company on Tuesday rolled out its Popular Dishes feature to help consumers decide what to order at a restaurant they're not familiar with. The new feature will recommend the most popular dishes at a certain restaurant based on "more than 100 million photos and reviews from the Yelp community," according to a Yelp spokesman.

Instead of awkwardly sitting at the table and scrolling through long comments on Yelp to figure out what to order, you can now find the feature at the top of each restaurant's page with a list of its popular dishes, photos and review snippets.

The company says it's using machine learning -- a buzzword, drawn from AI, among tech companies these days -- and its database of reviews and photos to make its picks.

Yelp isn't the first company to use machine learning in dish recommendations. UberEats launched a "recommendation for you" feature last year that drew on machine learning to customize the app based on what each person is ordering and favoriting.

"We trained a machine learning model to read through all of our restaurant reviews and recognize when users were writing about a dish they had ordered. We then match those dishes to other reviews and photos to find the most popular dishes at each restaurant," said Matt Geddie, product lead at Yelp. "We wanted to build this feature because we heard from users that they love using Yelp to figure out what to order, but we thought the process could be easier than looking through all of our great review and photo content individually."

First published on June 19, 12:52 p.m. PT.

Updates, 3:51 p.m. PT: Adds Matt Geddie statement.

Culture: Your hub for everything from film and television to music, comics, toys and sports.

Rebooting the Reef: CNET dives deep into how tech can help save Australia's Great Barrier Reef.