For the last 10 months, Carnegie Mellon University's Never-Ending Language Learning System, or NELL, has been continuously searching the web for text patterns and grouping them into different semantic categories, a system that closely mimics the way humans learn. But NELL has adopted another human behavior as well: tweeting everything she does.
When she learns a new fact, NELL adds it to her online database, and tweets her discoveries, so you can follow her progress. For example, today she learned that the phrase "dem franchise boys" fits in the category of "music artist." NELL's knowledge base currently consists of around 440,000 facts, with around 87 percent accuracy. Many of her tweeted lessons demonstrate amazing analytical capabilities – she was able to get "tool" from "Anchor Fixing Self Tapping Screw Bolts." The New York Times recently published an in-depth look at exactly how NELL learns these things.
NELL isn't perfect though. And thank goodness, because that means we get some pretty adorable not-quite-there-yet tweets, such as: "I think "chicken recipe time" is a #condiment" and "I think "anonymously" is a #fish." NELL's followers can tweet corrections to her and help her improve her associations. Check out her feed or browse her whole knowledge database to see more.
I think it would get more out of facebook.
By saying that NELL has "adopted" the human behaviour of tweeting you are misleading the reader. It is more likely that the software was specifically progremmed to do so and therefore has "adopted" no "human behavior". FAIL.
I'm curious about that statement too Tom91. It would be a great deal more startling if NELL "figured out" not only semantic categorization but set up "her" own Twitter account. SKYNET IS AWARE!!!
How can I tweet corrections to it? I found a few mistakes (like thinking that 'thick' is a colour), but there's very little info on how to rectify this.
I think that the researchers are calling all the tweet update shots ... otherwise they might get a hundred tweets come in making the bot think "A Boombox is not a toy."