Our intuitive AI outperforms (most) puny humans, claims MIT

Traditionally computers are great at crunching numbers, but lousy at understanding what they mean. But a team of researchers at the Massachusetts Institute of Technology thinks it's cracking that problem.

A new paper, to be presented at next week's IEEE International Conference on Data Science and Advanced Analytics, details the evolution of the university's Data Science Machine – a sort-of AI system that is adept at spotting trends and patterns in large chunks of data.

MIT ran the machine as a ringer in three human data science tournaments and had considerable success. Out of 906 human teams in the competition to find patterns in data fields, the computer system beat 615 of them.

The Data Science Machine managed to get within 87 and 96 per cent of the accurate answers submitted by human competitors. But, crucially, the Data Science Machine managed to do the job much faster than its fleshy competitors – human teams took weeks to divine patterns from the data while the computer took a maximum of 12 hours.

"We view the Data Science Machine as a natural complement to human intelligence," said Max Kanter, who built the Data Science Machine as part of his PhD.

"There's so much data out there to be analyzed. And right now it's just sitting there not doing anything. So maybe we can come up with a solution that will at least get us started on it; at least get us moving."

In the competitions, contestants were asked to examine large, relatively unstructured databases to look for contributory factors for functions like students dropping out of a course or the likelihood of a customer repeating a purchase at an ecommerce site.

The MIT team suggests that the computer system could be used to establish a baseline of results to check against a human response. As it gets refined, it may be possible to get extra insights by matching artificial and human operators against each other.

"The Data Science Machine is one of those unbelievable projects where applying cutting-edge research to solve practical problems opens an entirely new way of looking at the problem," said Margo Seltzer, professor of computer science at Harvard University who reviewed the study. "I think what they've done is going to become the standard quickly – very quickly."

In other words, if you're in information sciences, then it might be time to get worried about the competition. ®