[MUSIC PLAYING] JEN MORGAN: Hi, my
name is Jen Morgan, and I’m a systems
engineer here at SAS. Today, I’m going to show
you how Visual Analytics can help insurance companies
identify fraudulent claims. Today I’ll be investigating
37 open claims dating back several years to tropical storm
Bonnie, which made landfall in Southeast Florida. Three inches of rain and wind
gusts greater than 50 miles per hour were reported,
but damage was minimal. Yet in these open claims, we
have some high claim amounts. Let’s investigate these
claims using the geomap and cross tab features in
Visual Analytics Explorer. I’m going to plot my
claims for July 23, 2010 on a geographic map. So I need to select the
geographic map visualization. Next, I need to select the
homeowners address, the wind max speeds, and the
claim amount for my map. At the bottom of the screen,
you can see in the legend that the greener the bubble
the higher the wind speed, and the bigger the bubble
the higher the claim amount. I can see that my maximum wind
speed was 49 miles per hour and my highest claim
amount was $32,799. That seems odd to me. The property should not have
that much damage with that wind speed. So what I’d like to do next
is create a list report to view these 37 claims and
look at the contractor distance, look at the homeowner age,
review the homeowner income, as well as the claim amount. To do that, I’m going to
minimize this visualization and create a hierarchy so that
I can drill down into my data. I’ll select New hierarchy. And I first want to look at my
contractor, the type of damage that the home sustained,
and what the claim date was. Let me give it a
name and click OK. This hierarchy functionality
replaces the use of OLAP cubes. So what I can do
is select Crosstab to create a crosstab report. And I’m just going to drill
my contractor hierarchy right into the cross tab. I’m going to go ahead and
expand my information so that I can review it all. I’m going to select some
additional measures. I’d like to select the
claim amount, the homeowner age, the homeowner
income, and the distance that the contractor traveled. To do that, I can
come over to Measures. So I’ve got the claim amount,
I’ll add the homeowner age. Next, I’m going to
add the homeowner income and the distance that
the contractor traveled. In addition, I can
add totals to these. Interesting, the first
thing that I notice is that these claim dates are
more than a year and a half after tropical storm Bonnie
hit southeast Florida. I can also see that
the homeowners tended to be older on average, with
a higher than average income, and the contractors seem
to be coming from far away. This is really suspicious to me. And I’m thinking
that there could be some potential fraud on the
part of the contractors here. So what I’d like to do
is send these claims to the SIU unit for
further investigation. I hope you’ve enjoyed seeing
how Visual Analytics can help insurance companies
identify fraud. To learn more about
Visual Analytics, you can go to SAS.com/va.