White Collar Crime Risk Zones
April 25, 2017 8:32 AM   Subscribe

Modern life is fraught with perils, but thanks to The New Inquiry's new tool, you can know what your risk of being the victim of a financial crime is at any moment, anywhere in the US. Using state-of-the-art machine learning technology and predictive policing methods, combined with geospatial feature predictors and risk terrain modeling, you can see the risk to your livelihood presented block-by-block across the whole US. Using a database of people at a high risk to commit such crimes, the app even presents a generalized image of the potential perpetrator to allow you to be on the watch for anyone suspicious who may present a threat.

The full announcement, including a link to the white paper describing the methodology, is here.
posted by Copronymus (13 comments total) 22 users marked this as a favorite
 
Ha! This is next-level trolling. I have no idea if their model is accurate, but then I suspect the same could be said of "conventional" predictive policing models.
posted by Cash4Lead at 8:36 AM on April 25, 2017 [4 favorites]


line of best fit is not machine learning line of best fit is not machine learning line of best fit is not machine learning line of best fit is not machine learning etc
posted by Damienmce at 8:45 AM on April 25, 2017 [4 favorites]


Thatsthejoke.gif
posted by benzenedream at 8:58 AM on April 25, 2017 [5 favorites]


From this thread - alas because it's on gawker the comments are gone, best comment on an article evah:


"Whenever I see a group of white guys in suits walking down the street, I involuntarily clutch my stock portfolio statements a little more tightly."
posted by lalochezia at 8:58 AM on April 25, 2017 [24 favorites]


This is some fine trolling, but I'm vexed that it conflates being a victim of white collar crime with physical proximity to the crime. Like, I think a better troll would be to plot out where everyone who has been a victim of white collar crime lives and then watch as the map of the US lights up.
posted by Going To Maine at 9:04 AM on April 25, 2017 [5 favorites]


This is great.

Though, I agree with Going to Main - victim location seems like the right thing to use here rather than perpetrator location. I assume that was harder to find. Also, what the hell is "Enigma. (2017)" as a citation? That's some lazy bullshit scholarship that forces the reader to do extra work to figure out what the authors are talking about. Shame on them. But, still, I like it a whole lot.
posted by eotvos at 9:20 AM on April 25, 2017


I'm mostly completely intrigued by the local landmarks that are marked on that map. I had no idea that we had a Czecho Slovakian Association Hall, but apparently we do!
posted by ArbitraryAndCapricious at 9:21 AM on April 25, 2017


The next block over from me is owned in its entirely by a man whose legal name, shit you not, is "Richard Guy", of which he omits the middle syllable. I was hoping to see that block show up bright red.
posted by 7segment at 9:23 AM on April 25, 2017 [2 favorites]


I feel certain I've been sold securities by Richard Guy. Of course, I call him Dick.
posted by chavenet at 9:26 AM on April 25, 2017


Reminds me of this Larry Wilmore bit from the Daily Show.
posted by praemunire at 9:33 AM on April 25, 2017


This is hilarious, and the (fairly short) white paper is absolutely worth a read.
In developing our model, we focused on the physical features of the underlying landscape as described in the risk terrain modeling approach6 (explained further below), while deprioritizing temporal features. As such, we assert with a high degree of confidence that white collar crimes are occurring continuously in the predicted high-risk zones.
(My emphasis.) Brilliant.
posted by biogeo at 9:40 AM on April 25, 2017 [3 favorites]


More of this kind of trolling, less of the kind that revolves around yelling at women on Twitter.
posted by tobascodagama at 12:17 PM on April 25, 2017 [4 favorites]


The best part of the white paper is at the end.
As aproof of concept, we have downloaded the pictures of 7000 corporate executives whose LinkedIn profiles suggest they work for financial organizations, and then averaged their faces to produce generalized white collar criminal subjects unique to each high risk zone. Future efforts will allow us to predict white collar criminality through real-time facial analysis.

[ciswhitemale.jpg]
posted by AFABulous at 12:34 PM on April 25, 2017 [6 favorites]


« Older Shooting people is stupid. Guns are stupid. THAT...   |   The Philosemitic / Antisemitic Tchotchke Market Newer »


This thread has been archived and is closed to new comments