New research method predicts success, failure of D.C. restaurants

December 17, 2014 by Greg Muraski, University of Maryland

Using a database of 130,000 Yelp reviews of restaurants in Washington, D.C., two professors and a graduate student at the University of Maryland's Robert H. Smith School of Business have identified a method that allows software to "read" the content of those reviews—and predict which restaurants will close.

For the study, the researchers identified slightly more than 2,000 Washington, D.C., that were open as of December 2013. From various sources, they then identified roughly 450 that had closed from 2005 to 2014. To identify that foretold closure, they paired restaurants according to such factors as price and cuisine type, and looked at how the descriptions varied.

It doesn't take a Ph.D. to know that there's a connection between a restaurants' Yelp rating and whether it will survive. But what Jorge Mejia, a UMD doctoral student; Shawn Mankad, an assistant professor; and Anandasivam Gopal, an associate professor have created is more powerful: Their computer-assisted text analysis proved more accurate at predicting restaurants' demise than ratings alone. And it is most powerful when used in combination with numerical ratings.

"The whole idea is that we are surrounded by all of this free, unstructured data," Mankad says—hundreds of thousands of words that would require armies of employees to read, let alone interpret. "We should be using that data."

The influence of online reviews is indisputable: More than 60 percent of Americans say that such reviews have high or medium level of influence over their buying decisions.

Other scholars have sought to take the emotional "temperature" of online , by analyzing the proportion of positive versus negative words. This new approach goes deeper, examining constellations of words that were associated with restaurants' beating the long odds of their industry and remaining open.

For instance, restaurants for which reviewers used the words "food," "good," "place, "like," "order," "friend," "time," "great," "nice," and "service" tended to survive at unusually high rates. The Smith School professors called the variable linked to those "Quality_Overall," and it seemed to be the most potent signifier of general quality. "Constructing the variables, putting it into a predictive model—this is something that has never been done before," says Mankad.

They used one subset of data to uncover the relevant linguistic patterns and another subset to test the predictive power of their model. In that second group, the variables did predict, to a statistically significant degree, success or failure.

The working paper, "More Than Just Words: Using Latent Semantic Analysis in Online Reviews to Explain Restaurant Closures," grew out of Mejia's dissertation.

Mankad discussed the use of text data to improve business forecasting in October, at the Cornell Hospitality Research Summit. He also discussed the topic in a recent webinar sponsored by Hotel Management Magazine and the Wall Street Journal.

Explore further: Online food reviews reveal inner self, Stanford linguist finds

Related Stories

Economists find Yelp significant to restaurants

September 5, 2012

(—For anyone who has wondered whether crowd-sourced online reviews make a dent in a business's bottom line, the answer is an unequivocal yes, according to a new study by UC Berkeley economists.

Yelp acquires French rival Cityvox

October 28, 2014

The online review website Yelp announced Tuesday it had acquired French-based rival Cityvox in its latest overseas expansion.

A rainy day can ruin an online restaurant review

April 2, 2014

After looking at 1.1 million online reviews for 840,000 restaurants in more than 32,000 cities across the country, Georgia Tech and Yahoo Labs researchers have found that the weather outside can be just as significant a factor ...

Recommended for you

Nanoscale Lamb wave-driven motors in nonliquid environments

March 19, 2019

Light driven movement is challenging in nonliquid environments as micro-sized objects can experience strong dry adhesion to contact surfaces and resist movement. In a recent study, Jinsheng Lu and co-workers at the College ...

OSIRIS-REx reveals asteroid Bennu has big surprises

March 19, 2019

A NASA spacecraft that will return a sample of a near-Earth asteroid named Bennu to Earth in 2023 made the first-ever close-up observations of particle plumes erupting from an asteroid's surface. Bennu also revealed itself ...

The powerful meteor that no one saw (except satellites)

March 19, 2019

At precisely 11:48 am on December 18, 2018, a large space rock heading straight for Earth at a speed of 19 miles per second exploded into a vast ball of fire as it entered the atmosphere, 15.9 miles above the Bering Sea.

Revealing the rules behind virus scaffold construction

March 19, 2019

A team of researchers including Northwestern Engineering faculty has expanded the understanding of how virus shells self-assemble, an important step toward developing techniques that use viruses as vehicles to deliver targeted ...

Levitating objects with light

March 19, 2019

Researchers at Caltech have designed a way to levitate and propel objects using only light, by creating specific nanoscale patterning on the objects' surfaces.


Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.