GUY RAZ, host:
From eclipsing stars now to a computer program that could make sportswriters disappear. It's the product of the Intelligent Information Laboratory at Northwestern University, and it's called StatsMonkey. What it does is take the statistics from a baseball game and produce a computer-generated news story about that game.
Dr. KRISTIAN HAMMOND (Co-Director, Intelligent Information Laboratory, Northwestern University): University Park: An outstanding effort by Willie Argo carried the Illini to an 11-5 victory over the Nittany Lions on Saturday at Medlar Field. Argo blasted two home runs...
RAZ: That's Kris Hammond. He is co-director of the lab at Northwestern, and he's reading a story generated by the StatsMonkey program.
And I have to say, Kris Hammond, that sounds pretty close to what you might read in the local paper.
Dr. HAMMOND: It's - I mean, that's what it's aimed at being.
RAZ: So how does it work? How does the program work?
Dr. HAMMOND: Well, it starts with the numbers. And, in fact, in general, what we do is we go from numbers to story. So it looks at the box scores, it looks at the play-by-play information. And then it uses that to figure out what we call the angle. That is, what kind of game was this? Was it a back and forth? Was it a pitcher's duel? And then from that, it actually generates the language.
RAZ: What if something sort of a bit unusual happens, you know, say, a pitcher throws a no-hitter or, you know, a bat slips out of a hitter's hands and injures a fan? Can the program actually handle those scenarios?
Dr. HAMMOND: Yes and no. For a no-hitter, absolutely. For a bat slipping, no because that's not going to be reflected in the numbers.
RAZ: Why - I mean, you work at Northwestern, at the journalism school at Northwestern. Why would a journalism school sort of be interested in creating a computer program that could eventually make what you do and the people you train irrelevant?
Dr. HAMMOND: That's a good question.
(Soundbite of laughter)
Dr. HAMMOND: It doesn't make anybody irrelevant. We're really aiming this at a genuinely local audience. That is, we're trying to write the stories that no one else is writing.
We could literally write a game story for every single Little League game that's played in this country. And that means every kid, every dad, every family, every grandma, would see the story of what their kid is doing.
RAZ: At some point, you're going to lose control over that technology. And so, I'm wondering if you don't see it as a technology that will replace human beings as journalists?
Dr. HAMMOND: Yes, someday, we'll build something that can do absolutely everything a journalist can do, given that they - you know, the facts are in front of them. But that's our goal. Our goal is to genuinely model human thought, intelligence, reasoning, but always looking to the long tale, looking to the places that are underrepresented, underreported, and trying to fill that niche.
RAZ: What about sportswriters? Have any of them contacted you and said, you know, cease and...
Dr. HAMMOND: Oh, absolutely.
RAZ: They have?
Dr. HAMMOND: Not cease and desist.
Dr. HAMMOND: We actually had a writer who did a side-by-side of his story and our story. And he really said - he said, I hope you guys don't try to replace me because I like going to games, and it would make me sad if that was no longer part of my job. And honest to God, we have no interest in doing that.
RAZ: So no chance of, you know, I'll be replaced by a computer anytime soon?
Dr. HAMMOND: In a couple weeks.
(Soundbite of laughter)
Dr. HAMMOND: No. Not anytime soon.
RAZ: Kris Hammond is co-director of the Intelligent Information Laboratory at Northwestern University. He joined us from member station WBEZ in Chicago.
Kris Hammond, thanks so much.
Dr. HAMMOND: Thank you.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.