Skip to Content

Every Spreadsheet Has a Narrative to Tell—Just Add Some AI

To make sense of data, we are teaching computers to speak our language.
November 7, 2017
Justin Saglio

Every day, we export 2.5 quintillion bytes of data out of our ubiquitous machines. This data comes in forms from million-row spreadsheets to photos of local coffee shops. It is filled with information, but most of it is not naturally in a form humans can understand. Kris Hammond, chief scientist and cofounder of Narrative Science, is transforming this data into language.

“Language by itself is miraculous and uniquely human,” Hammond told the audience at EmTech MIT on Tuesday. “You can teach a dog things; crows use tools; beavers use dams. There is no other creature that uses language the way we use language. Although machines use words, they struggle with language.”

Humans innately connect complex ideas in our heads, Hammond said, because of our experience with language. Enabling machines to communicate using the same language as we do becomes ever more critical as machines become a larger part of our lives.

Narrative Science is using AI to teach machines this human skill. Its software, Quill, can take data like the box score of a baseball game, summarize the content, and extract a “narrative” from it (see “Who Will Own the Robots?”). Now, instead of staring at a bewildering compilation of numbers, you can view an easy-to-read paragraph telling you what you need to know.

“We have struggled with data for a long time. We shouldn’t be struggling with it; the machine should be presenting it to us,” Hammond said.

The city of Chicago has taken this approach to synthesize data about miles of its shorelines. It records data on numerous beaches and creates cumbersome spreadsheets. Using Quill, the city boils the information in the spreadsheets down to digestible statements indicating the best and worst beaches in the city.

Turning data into language can also benefit businesses and their customers.

“Most people who have jobs don’t also want to look at data and find out what’s going on,” Hammond said. “You don’t want to hand data to people who run the deli. You take that exact same data and turn it into reports to give them advice about what they might be able to do next.”

Keep Reading

Most Popular

light and shadow on floor
light and shadow on floor

How Facebook and Google fund global misinformation

The tech giants are paying millions of dollars to the operators of clickbait pages, bankrolling the deterioration of information ecosystems around the world.

This new startup has built a record-breaking 256-qubit quantum computer

QuEra Computing, launched by physicists at Harvard and MIT, is trying a different quantum approach to tackle impossibly hard computational tasks.

wet market selling fish
wet market selling fish

This scientist now believes covid started in Wuhan’s wet market. Here’s why.

How a veteran virologist found fresh evidence to back up the theory that covid jumped from animals to humans in a notorious Chinese market—rather than emerged from a lab leak.

protein structures
protein structures

DeepMind says it will release the structure of every protein known to science

The company has already used its protein-folding AI, AlphaFold, to generate structures for the human proteome, as well as yeast, fruit flies, mice, and more.

Stay connected

Illustration by Rose WongIllustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.