Email

The internet is a powerful tool, which allows billions of people from around the world to share ideas. Nothing like it has ever existed, and new ways to tap into this incredible resource are developing on a daily basis. One of the latest being crowdsourcing.

Crowdsourcing is a way of using the internet to access the collective opinions and knowledge of internet users around the globe. As of Friday, July 15, 2011, Applied Research Associates, Inc are tapping into this resource with a program called Aggregative Contingent Estimation System (ACES).

This program invites users to make their best predictions on various intelligence issues.

“We’re trying to make good use of everybody’s individual opinions and trying to determine what aspects of them might be important and what would lead to a good forecast,” said Applied Research Associate’s principal investigator of the ACES project, Dr. Dirk Warnaar.

Once it is up and running, the system will narrow in on populations with the most accurate predictions, and possibly use the information for security and intelligence purposes, largely depending on the success of the program.

The Applied Research Associate program is already working on new features to improve the user experience and accuracy of the predictions. The feature being developed will track user’s predictions and the accuracy of the results to provide feedback to the individual about his or her predictions.

For example, if a person consistently guesses that a certain event has a 70% chance of occurring, but the data shows that a majority of the time, there is only a 45% chance, the program will notify the user of this data. This information will hopefully be used to improve the accuracy of predictions for each user.

The program may implement a social element to providing predictions. One possibility on the table is to ask participants to provide elaborations on why they predicted a certain event, and allowing other users to vote on the prediction.

“We think there’s a way of doing it this way that would basically give us the arguments for [a prediction] and split the problem apart in such a way that people start thinking about this problem a little deeper than they otherwise would,” said Dr. Warnaar.

There is some fear that including a social element may deter the accuracy of predictions by introducing group think, making users less likely to provide their true opinions.

Article Written by Marisa O’Connor