This Article Nails Natural Language Generation, What I Do With It, and Why I Do It


My parents still don't really know what I do.

I mean, they know I'm at Automated Insights and they know I've been there from the beginning and do my part to run the place. They know I do those fantasy football articles (well, that's not really me anymore but my fingerprints are still all over it). They know I use computers to create words out of data.

But they're not really sure how. Or why. Or why anyone would pay me to do that.

A bunch of my friends fall into the same bucket.

So I've been fighting this battle for the last six years, with sometimes limited and sometimes satisfying results. When we landed the fantasy football recaps (we do this for the two biggest providers), a light went on for a lot of people. Oh -- you take the scores from everyone's fantasy football matchup and use computers to write millions of recaps in a couple hours. Makes fantasy seem more real and more fun.


But that's a minor scratch on an Earth-sized surface.

Last week, I sat down for an interview with Mike Riggs from FreeThink, and we talked for close to an hour about Natural Language Generation, Automated Insights, fantasy football, the Associated Press, and a bunch of other stuff that has made up my six years in the science of automated content.

I don't know if it was that Mike was just the right amount of prepared versus curious, or he's just a likable person and a good interviewer, or maybe Wednesday happened to be a really good and clear-minded day for me, but it all kind of flowed out, and Mike did an excellent job piecing it together -- leaving in the insights and weeding out the chatter.

Which is exactly what I try to make the computers do.

So whether you know me or not, if you've ever wondered what Natural Language Generation or automated content is and why it's such a big deal, or why it's not going to put you or anyone else out of a job, and where all this machine learning and artificial intelligence and smart robot writing is headed -- read this article and let me know what you think.

At the very least, you should watch the livestream video of two of our employees playing a ping pong match with color commentary provided by our Wordsmith NLG product and spoken by Alexa. That video killed when I brought it to my talk at South by Southwest earlier this year, and it hadn't been made public until this article.

read the published article at:
Let me know what you thought about this article, good or bad.

Your Email Address

Leave it blank to be anonymous

Your Thoughts?

Defining the Context Layer in NLG Part 2: What Does It Mean?
Good NLG leaves the reader with actionable information—and the creator should write to that.

Defining the Context Layer in NLG. Part 1: What Just Happened
In this post, we'll look at the first and most important structural elements, and talk about building a context layer around it.

That Old Adage About a Startup With a Solution Looking for a Problem
In Episode 6.7 of The Startup Show, we talk to NeuroPlus founder Jake Stauch about his amazing brainwave tech.

When a Startup Takes On Big Corporations
In Episode 6.6 of The Startup Show, we discussed the challenges of starting up a retail product.

The Fine Line Between Love and Hate in Startup
Last week, we discussed whether or not Teaching Startup should have a future.