3 ms·

Robo-journalism: How a computer describes a sports match

3 points by SimplyUseless 11y ago

ClintEhrlich 11y ago
The last few years, the ability of software to generate convincing natural-language prose has become more and more impressive. I first recognized the technology in my life when there was an earthquake last year and the LA Times' website instantly posted a bare bones computer-generated article, far faster than any human could type. At the time, I was not particularly impressed. It seemed like the software had been working from a prewritten template into which it plugged in details like the location, timing, and intensity of the quake. The program in this article is more sophisticated, because it looks for patterns in the data in order to determine which facts should be included in the article. But it sounds like most of the pattern-matching involves simple, prewritten, domain-specific rules. Some person had to make a list of as many basketball concepts as possible and then write functions for identifying them in the box-score. The result is cool, but the program doesn't demonstrate that much more intelligence than the Derrida generator. What would be really cool is if someone figured out how to automate the creation of the robo-journalist's conceptual ontology. I don't think that would be an AI-complete problem, as long as the corpus was limited enough. For sports, articles about baseball games would probably be the easiest place to start, because there are so many games, each of which involve a predictable, linear, number-oriented narrative.
- waterlesscloud 11y ago
  I fiddled with a robo-narrator for baseball earlier this year. MLB games are nice to work with as the league provides game data in a very detailed, very easy to parse format, updated live as the game progresses. It was just a toy project to help me learn some Racket. I had it hooked into the windows voice synth api, which was neat to listen to. But all it really did was a lot like your basketball idea- look for some typical patterns (which are common in baseball) and link them together. I had more ambitious thoughts about it, but wandered off to something else. :-)
- schoen 11y ago
  It seems like quite a broad range of "exceptions" can also occur in the baseball game. For example, one side can protest the umpire's actions, or a player can be injured (the nature and consequences of the injury should be described -- the linked article gives an example of this!), or the game can be suspended as a result of weather or a natural disaster, or something else can interfere with the continuation of the game (like fans running onto the field, or a brawl). Also, some very unusual events within the ordinary course of the game (like going into many, many extra innings) might call for a description of how people reacted to the situation. (And some sports journalism would try to describe players' presumed intent in doing something unusual, like a sacrifice play, or the manager's motivations for making a substitution.)

3 points by SimplyUseless 11y ago

The last few years, the ability of software to generate convincing natural-language prose has become more and more impressive. I first recognized the technology in my life when there was an earthquake last year and the LA Times' website instantly posted a bare bones computer-generated article, far faster than any human could type. At the time, I was not particularly impressed. It seemed like the software had been working from a prewritten template into which it plugged in details like the location, timing, and intensity of the quake. The program in this article is more sophisticated, because it looks for patterns in the data in order to determine which facts should be included in the article. But it sounds like most of the pattern-matching involves simple, prewritten, domain-specific rules. Some person had to make a list of as many basketball concepts as possible and then write functions for identifying them in the box-score. The result is cool, but the program doesn't demonstrate that much more intelligence than the Derrida generator. What would be really cool is if someone figured out how to automate the creation of the robo-journalist's conceptual ontology. I don't think that would be an AI-complete problem, as long as the corpus was limited enough. For sports, articles about baseball games would probably be the easiest place to start, because there are so many games, each of which involve a predictable, linear, number-oriented narrative.

waterlesscloud 11y ago
I fiddled with a robo-narrator for baseball earlier this year. MLB games are nice to work with as the league provides game data in a very detailed, very easy to parse format, updated live as the game progresses. It was just a toy project to help me learn some Racket. I had it hooked into the windows voice synth api, which was neat to listen to. But all it really did was a lot like your basketball idea- look for some typical patterns (which are common in baseball) and link them together. I had more ambitious thoughts about it, but wandered off to something else. :-)
schoen 11y ago
It seems like quite a broad range of "exceptions" can also occur in the baseball game. For example, one side can protest the umpire's actions, or a player can be injured (the nature and consequences of the injury should be described -- the linked article gives an example of this!), or the game can be suspended as a result of weather or a natural disaster, or something else can interfere with the continuation of the game (like fans running onto the field, or a brawl). Also, some very unusual events within the ordinary course of the game (like going into many, many extra innings) might call for a description of how people reacted to the situation. (And some sports journalism would try to describe players' presumed intent in doing something unusual, like a sacrifice play, or the manager's motivations for making a substitution.)