In this interview with Matt Gould, CSO of ARRIA NLG which is a platform that generates natural language to create data storytelling, we begin by discussing the nature of how we see, and how the natural chaos of “seeing” is organized through the development of language skills. Structure and syntax is fundamentally the same throughout the world,  no matter what specific written or spoken language we use: there is “beginning, middle, and end”, a story form that language adheres to, regardless of locale.

Converting data into data storytelling through Arria's Natural Language Generation; a weather report

Converting data into data storytelling through Arria’s Natural Language Generation; a weather report

This premise forms the foundation of the distinction of the Arria NLG Platform, a form of artificial intelligence software that specializes in extracting information from complex data sources and communicating that information in natural language.

Arria’s scientists have authored over 300 academic papers on the subject, and their senior scientists wrote the de facto textbook in the field, ‘Building Natural Language Generation Systems’.

While even the cleverest or most obvious visualization of data follows the rules of storytelling of some kind, if the platform can be instructed on not only how the data is to be shown but how to tell that story through words, even to different audiences with different levels of interest or understanding, you can give that data (or, rather, the analysis and the meaning behind the data) its own voice with great clarity and a better ability to communicate. Whereas infographics and interactive data visualizations are human-made outputs that tell the story behind data (and typically in only one way) this platform learns the way data is to be analyzed for particular uses, and how to tell the story behind it for the different consumers of that information or knowledge. It then automates the reporting through a combination of data visualizations and the story (in text) to fully explain its meaning.

After explaining the process Matt demonstrates Arria’s platform using examples including: written weather reports generated by the platform in seconds in natural language; three different ways of reporting on a newborn’s health, again generated in moments, each report using the language of (respectively) the attending physician, the nurse, and the newborn’s parents; and business analysis that include not only reporting on trends and existing information, but creating predictive analysis and suggestions based on business practices and interests taught to the system—all, again, in natural language for the audience.

Arria NLG also offers its technology as pre-packaged SaaS Products and as a Software Development Kit (SDK) with APIs that allow developers to add NLG functionality to their own applications.