Dysrupt Labs – Improving event prediction using collective intelligence

Since inception in 2013, the Dysrupt Labs team have developed a human-centric platform that harnesses collective intelligence. The platform helps organisations access more accurate and timely forecasts compared to traditional forecasting methods.

“Prediction markets are accurate about 7 times in 10. If they are lucky, experts are right about 6 times out of 10. We have taken the prediction markets and using artificial intelligence have increased the accuracy from 7 to 8 times out of 10,” Mr Mattingly said.

Through collaboration with the Brain Mind and Markets Lab at the University of Melbourne, Dysrupt Labs has applied artificial intelligence (AI) to the dataset to increase accuracy. The platform has been built to use AI to identify when a poll or a survey or market is possibly wrong. AI takes out the emotion of the crowd to provide a more measured view of opinion.

Mr Mattingly says Dysrupt Labs analysed the 2016 US election using their platform. “The world said that Hillary was 87% more likely to be president the day before the election. Our AI said that she was 53% likely and that Trump's election was in the margin of error. In other words, AI predicted that it was a coin toss when the rest of the world said it was a landslide.”

To date the research carried out by Dysrupt Labs has been focused on re-engineering, re-imagining and re-conceiving existing prediction markets to improve forecasting accuracy. The company’s research also uses combinatorics, a mathematical theory relating to the likelihood of a combination of events happening.

The target market for commercialisation is government, hedge funds and banks. Governments can use this sort of intelligence for geo-political forecasting.

Hedge funds and banks can identify key drivers of interest rates, inflation rates, exchange rates and listed equities. The team is also exploring potential applications in medical research, disease and epidemic tracking.

Dysrupt Labs filed a joint patent in February 2018 with Murdoch Childrens' Research Institute at the Royal Childrens' Hospital in Melbourne.

The R&DTI also allowed the Dysrupt Labs to build capability in Melbourne, including skills in financial software design, price formation in prediction markets and the detection of expert traders. The company has a team of 12 working on the platform, most of whom are mathematicians.

To gain entrance in to the R&DTI, Dysrupt Labs initially engaged The Compliance Group to help them apply for the program. This proved a successful way to quickly understand what the program was, how to apply and what was required to maintain appropriate records. Since then, the team has built in-house expertise and The Compliance Group assist in an auditor role.