How self-learning computing changes government decision making

article-body

Tom: Kerry this question or area of cognitive computing,  I think IBM Watson is your product or solution in that space, what is that and how do agencies think about self-learning cognitive computing?

Kerry: Look I’m going to try to keep it as simple as possible. Watson is as you say it has cognitive computing capabilities. It’s computers that learn. So the more you feed it, the smarter it gets. So the medical institutes, the institutions we’re working with using are feeding more and more data into Watson to allow them to make smarter decisions, more accurately, and at a much faster rate than they were ever been able to do before.

So we’ve talked about here in Australia — it’s public — with Boarder Protection, we’re working with Boarder Protection on trialing, (now) we’re moving from trial basis to a more formal engagement around identifying potential threats and potential areas of risk for Australia Border Protection. Drawing insights again from what we call this unstructured data, from stuff that’s just out there floating around.

Tom: And the point being all done in relatively real time?

Kerry: Absolutely.

Tom: Before you would have to truck it off to somewhere, analyze it, bring it back.

Kerry: It would have taken months and in some cases years to draw the same level of insights. So the speed and then the accuracy, to enable people to make more informed decisions on what we call and we’re very passionate about, the actionable insights. So it’s not just stuff that, it will get a lot of data. So what is the stuff that we’re going to go and action that will deliver a better outcome for Australian citizens at a federal or a state level.

Tom: And I think for agencies, the insights are very powerful tools now. It allows them to have these many touchpoints with their citizens and be able to follow that in real time and as you say, get real insight to be able to manage that better on behalf of the citizens and consumers so their experience is better.