Software paraphrases sentences

Key to the technique is comparing news sources that cover the same events but employ slightly different styles. Because they are writing about the same events they contain the same facts, or arguments, said Barzilay. “This gives us patterns which are kind of the same — and this is the core of the paraphrasing technique.”…. [T]he system learned incorrectly that “Palestinian suicide bomber” and “suicide bomber” were the same, and that “killing 20 people” is the same as “killing 20 Israelis”, said Barzilay. These mistakes made by the system are “due to how reporters are reporting,” she said. “In some sense… the teacher here is what the reporter writes,” she said. Kimberly Patch —Software paraphrases sentences (TRN)

The Palestine/Israel detail is presented as an example of pro-Israel reporter bias, but I’m not so sure. If, according to the sample of news reportage being examined, more Israelis were killed than Palestinians, and if the ways in which Israelis were killed (civilians killed in marketplaces by suicide bombers, and also soldiers killed by armed combatants) was more newsworthy than the ways in which Palestinians were killed (armed combatants killed by soldiers and some innocent bystanders killed by soldiers) then the computer’s “mistake” might be understandable. But I’m not informed enough about the research involved to be able to make any reliable statement; of course the computer isn’t responding to what really happened in the world, it’s responding to the way a certain group of reporters described what their research tells them happened in the world. Of course, the results are going to reflect human biases, but the sample fed into the computer is affected by such things as how likely a news source that reflects a particular worldview will publish an online English edition.

On a lighter note…

Speaking at a press conference, researchers shifted uncomfortably from foot to foot and coughed into their hands before insisting, “Of course this software won’t be marketed to students intending to fool turnitit.com. Whatever gave you that idea?”