SoylentNews is people
SoylentNews
Submitted via IRC for AndyTheAbsurd
Google Street View has become a surprisingly useful way to learn about the world without stepping into it. People use it to plan journeys, to explore holiday destinations, and to virtually stalk friends and enemies alike.
But researchers have found more insidious uses. In 2017 a team of researchers used the images to study the distribution of car types in the US and then used that data to determine the demographic makeup of the country. It turns out that the car you drive is a surprisingly reliable proxy for your income level, your education, your occupation, and even the way you vote in elections.
Now a different group has gone even further. Łukasz Kidziński at Stanford University in California and Kinga Kita-Wojciechowska at the University of Warsaw in Poland have used Street View images of people's houses to determine how likely they are to be involved in a car accident. That's valuable information that an insurance company could use to set premiums.
The result raises important questions about the way personal information can leak from seemingly innocent data sets and whether organizations should be able to use it for commercial purposes.
(Score: 3, Informative) by Anonymous Coward on Sunday May 05 2019, @03:42PM (3 children)
Reading the study, the researchers looked at 7 variables found in each picture of a residence. 5 of them were found to improve the statistics.
Neighborhood type
Street View quality (i.e. the quality of the image Google took)
House Type
House Age
House Condition
"Building Density" and "Wealth of residents", were the other two and did not improve the statistics. My theory is that these are already captured by the insurance company's models.
So, how does one improve their situation so that the model produces lower risk profiles for them? By improving those 5 variables (at least in Poland). The problem is, there isn't much a person can do to change 4 of those. Only the "House Condition" is a variable a person can reasonably change. Of course a person could also move to a house that has better scores on the variables. That poses some questions across several dimensions; affordability, urban/rural distribution, traffic/congestion, racial factors, socio-economic factors and many more.
The study: https://arxiv.org/ftp/arxiv/papers/1904/1904.05270.pdf [arxiv.org]
(Score: 4, Touché) by maxwell demon on Sunday May 05 2019, @05:18PM (2 children)
I wonder what causes the correlation between image quality and car accident risk (and if higher image quality means a higher or a lower risk).
The Tao of math: The numbers you can count are not the real numbers.
(Score: 2) by Farkus888 on Sunday May 05 2019, @06:26PM (1 child)
True correlation, as in not Texas Sharpshooter style P hacking, means one of 3 things. X causes Y, Y causes X or X and Y are both influenced by some third variable Z. Walk through a trailer park and count dents per 100 vehicles and compare it to one of those developments that has a stone sign out front. This is such an obvious gimme to anyone who hasn't spent their entire life in one economic class that I'm surprised the insurance companies hadn't already controlled for it. Applying this data creates a disincentive to buy a dumpy house and fix it. Those people are a little crazy anyway.
(Score: 1, Insightful) by Anonymous Coward on Sunday May 05 2019, @07:01PM
They pretty much already did. Crime rates, crime type, and zip code show a lot more about what is going on. Because guess what poor people live in cheap areas and rich people live in rich areas.
The SJW type has decided that also correlates to the color of your skin or nationality. Not because of 'racism' but because social mobility is tough with out greenbacks. Hell I moved a couple of years ago from one city to another and that was tough as balls. I was moving around in the same social strata! With no money you are moving pretty much nowhere. So they have switched up metrics. But those metrics are just fungible substitutes for the ones they have deemed racist. So yeah insurance companies and banks already had it figured out. They both talk about 'risk'. They use whatever metrics they can to figure out how much you make and likelyhood of payback all to see if they can become middlemen in any deals you make to get a cut.