Wednesday 31 August 2011

Using Crowdsourcing to Predict Terrorist Threats

New ways are constantly sought to better predict and tackle terrorist threats, and thanks to the advent of social technology, crowdsourcing could potentially play an important role. Research is being carried out by a team of academics, Applied Research Associates (ARA) based in Raleigh North Carolina to determine how crowdsourcing can be used to collect data hence make accurate predictions and spot trends.

Working in partnership with seven universities around the US, the ARA are testing and developing technology systems that can efficiently collect and process different opinions on topics such as politics, the economy and disease control. This information is then passed onto the intelligence agencies to help them make predictions.

The ARA have considerable experience in the field of threat assessment having already developed the Federal Security Risk (FSR) manager. This sophisticated software carries out detailed threat assessments examining likely attack scenarios and analysing possible vulnerabilities. FSR is regularly used to carry out assessments for military facilities, government offices, court houses and industrial estates.

For its latest venture the ARA are using Forecasting Ace, a platform still in beta stage. The general public are invited to register and then render their opinion on any subject matter giving details as to their reasoning. It also allows people to include additional information such as their background and expertise be it politics, technical, academic, or industrial fields. 
Various crowdsourcing have been used in the past. In general they take an average of everyone's opinion and come up with predictions. But there is very little research on whether these methods work. We think we can do better by weighting people's opinion based on their rational for making their predictions. - Dick Warnaar, Principle Project Investigator ARA 
Whilst the concentration is on predicting possible terrorist threats, the ARA team hope to use it to predict other outcomes such as fuel and energy prices, and performances on the stock market.

Image Credit; Josh Satiger

No comments:

Post a Comment