Big data analysis is essential to our information age as data pile up and too few people can properly analyze them and find hidden algorithms that have some kind of predictive power. Until now, finding those algorithms required something that only humans had – intuition. But the Massachusetts Institute of Technology team hope that their machine may soon make human intuition obsolete.
The team already tested their new system in three separate competitions. About 900 human teams took part and the MIT’s Data Science Machine outperformed 615 teams.
In the first two competitions, the forecasts made by the system had a 94 percent, respectively 96 percent accuracy. In the third competition accuracy fell to 87 percent, which is still an impressive feat for a non-human system. Yet human teams worked on their predictions for months, while the machine needed at most 12 hours to provide an accurate answer.
“We view the Data Science Machine as a natural complement to human intelligence,”
said Max Kanter, a MIT researcher whose academic work helped the team build the machine.
Kanter also said that there are heaps of unchecked data which humans have yet failed to analyses. A computer system like the Data Science Machine may help computer scientists at least start analyzing big data.
Kalyan Veeramachaneni of the MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) recently stated that MIT’s Data Science Machine may soon replace humans in big data analysis, as the competitions’ results showed.
Veeramachaneni explained that the new system does not only find hidden patterns in big chunks of unfamiliar data, but it can also solve problems related to big data. For instance, it can predict what the chances are for a student to drop out an online course.
Scientists explained that their machine needed two ‘variables’ to make its predictions. It first looked at how much time the student spent on the online course. Second, it analyzed just how much time before a deadline to submit an assignment the student began working on the assignment. The MIT team explained that the institute’s site does not collect such data but it does collect data from which the statistics can be inferred.
Margo Seltzer, a computer science expert at Harvard University was impressed with the machine. Seltzer said that at this pace, machine analysis of big data may “become the standard” real soon.
Image Source: Pixabay
Latest posts by Alan O’Leary (see all)
- Woman Found Alive After Missing for 42 Years - Oct 30, 2017
- October Will Welcome The Draconid Meteor Shower And The Orionids - Oct 6, 2017
- Scientists Are At A Loss After Unearthing A Porpoise Grave - Sep 22, 2017