With prediction tools and apps being all the rage today, their reliability and accuracy are not always there. Before the word “prediction” took over, we used to have words like “trending” and number crunches used for a lot of what we saw, but today it’s the big buzz word “prediction”. Machine learning algorithms certainly can be useful but they too make errors. A while back I connected data selling to lack of privacy and it got some attention, but I think not enough.
Today as companies see revenue streams disappear from their line of business, they turn to the easy solution of buying and selling consumer data, but it just doesn’t end there, they have to do something with it to create value, and that line of business today is called “scoring”. It’s happening all around you and it’s evolved to what I call “Excess Scoring” as you have data scientists cranking and wrangling statistics and data every which way but loose to find out how to find some data with queries to sell. Bloomberg bought a data mining company which they intend to use in house and also sell to their clients so they can predict the behavior of clients. Bloomberg of course might be motivated as they stand to lose a lot of money when JP Morgan and Bank of America cut thousands of Bloomberg terminals which rent for $21,000 a year, which has been the life blood of the financial world for years. Replace that revenue, mine and sell data.
Bloomberg is only one of many examples out there too as you also see IBM pushing this cognitive data mining like crazy, and that’s what 13 quarter of declining sales and too many years of stock buybacks end up producing, where do you make up the lost revenue, mining and selling consumer data. Sure there are some value areas with mining big data when it comes to “real” research like some in healthcare, but even some of that gets blurred as you end up being marketed to believe some app is a diagnosis tool, when in fact it’s nothing more than another risk assessment that gathers, mines and scores you and sells the data and the scores.
Many healthcare companies get around the HIPAA areas with in fact not selling your data that they have, but rather just selling a “score” and there’s little HIPAA help there and with that, so goes any privacy. When computer scientists and Quants are under pressure to create a revenue producing model, things go a bit crazy and correlations become spurious and data becomes flawed.
So what happens when the data becomes flawed? Are there avenues for consumers to use to correct the flawed data and scores being produced and sold for money, which by the way is a $180 Billion dollar a year business? Few and limited. Argus Analytics is one firm that buys up credit card data and puts out tons of “scores” about your behavior patterns. How many of you have heard of that company? Probably not many I would guess.
Argus Analytics Produces “Share of Credit Card” Data On Consumers - Digs Up The Dirt on Your Credit Card Behavior Patterns-US Consumer Protection Agency Is A Client-We Are Paying for Richard Cordray’s Slow Education Process
There have been some very faint efforts in Congress, they spend a lot of time talking about it, but again you have lawyers selling pixie dust that some magical new algorithms are going to protect consumers and on the deception goes. With indexing and licensing data sellers, at least we would know who they are and where go to corrected flawed data. Do we have that now? No! You can find some of them but not all, especially the obscure companies that are taking the data and “scoring” you and selling those results.
BDEX is a “scoring” and data selling company, you probably never heard of. Take look over there and see what’s going on with “scoring” you and your data. They’re getting everything they can get their hands on and also tech and promote (and make money)showing others how to score you as well. You’ll find Live Ramp from Acxiom over there too as a partner, Acxiom one of the biggest data brokers around.
Most people have no clue that all their electronic transactions are for sale, there’s no privacy so use cash when you can if you want a little bit of privacy. It’s all over the place and read the link below on now how CMS wants to sell Medicare data (link below). Andy Slavitt, former director at CMS is a big promoter of “scoring everything and everybody” with a background of being a former Goldman banker and a high ranking United Healthcare Executive that racked everyone over the coals. His big claim to fame was the error producing Ingenix Algorithms with physician pay and then later with medication adherence predictions with Express Scripts, selling algos to them. The lawsuits still continue. CMS wants to “approve” those who get the chance to buy and sell your Medicare data though, but we know how that goes, no regulation there as all as again, we don’t know who all the data sellers are in the US and exactly what they do?
Unfortunately Data Scores get used frequently out of context and you have no idea how many times and how you have been scored. When it comes to skewing a correlation to make money, data scientists and quants will stop at nothing if it makes money as their job may depend on it. So in fairness there’s some outside pressure to keep a job that might make them cheat a bit and go spurious.
The World Privacy Forum was all over this a couple years ago with their extensive report “The Secret Scoring of America” and it’s a good report. In virtual worlds everyone gets to be LeBron, but in the real world, there’s only one LeBron. We can thank Larry Ellison for those words of wisdom as sure he sees it loud and clear as well. He’s a software engineer by trade and people that write code and create software are all very much aware of the bad side, but choose whether or not to talk about it. Take note of the Alter egos of software engineers, data scientists and anyone who codes for big profits, i.e. Facebook.
World Privacy Forum Report - The Scoring of America: How Secret Consumer Scores Threaten Your Privacy and Your Future - One Big Element that Fuels the Continued Attack of Killer Algorithms & Demise of the Middle Class Creating Profiteering And/Or Denial of Access
As consumers it’s only right that we at least know who’s scoring us and using our data when it gets flawed. Are we to live out lives out with flawed profile created about us for the sake of corporations making big money selling it? Why even get out of bed every day when we are “scored” to the hilt with risk assessments that are like a ball and chain around our ankles I ask?
Sure there’s some useful scoring going on out there and we need some of that but there’s a lot of it that’s gone over the top. Even UCLA this week came out and said they suggested to stop using the BMI to determine a person’s over all health. Data Scorers can take that number and wrangle it into a massive amount of other data bases and turn you into a huge “risk” assessment that shouldn’t be allowed to walk on this earth! That’s where we are getting today as more and more bogus risk assessments are created that are nothing more than algorithms and math models to make money.
Step one for any real privacy to exist is to determine who all the data sellers are and what they sell. I’ve had honest quants and mathematicians alike as well as the World Privacy Forum agree with this as a first step of action versus the whacked out perceptions that are being promoted out there now as privacy efforts. How about when you get (your data) gets repackaged and sold? None of the privacy dupes are talking about that and it does exist. If you don’t think this exists, you have been spending too much time living in some virtual worlds out there that exist. Look at the recent case with Uber and Unroll.me.
Anthem Data Breach–Crooks Want to Sell Data Too–The Impact of the “Data Selling For Profit Epidemic” That Exists in the US With Scoring Consumers Removing Access and Human Dignity
Prescription medication adherence scoring turns you into a non compliant Outlier and that data gets sold all over the place. Most have no clue on how much money gets made selling your prescription data. If you pay cash and they can’t find enough credit card data to prove you are filling a prescription, you default to being a non compliant Outlier. Nice, huh? How do you fix that? These scores are all secret and you can’t get them, so you now there’s a double whammy, secret scores and secret data sellers you have no access to, can’t fix anything as a consumer that corporations “score” about you for a profile.
The post below was generated by a conversation I had with a real pharmacist with these concerns. Read it and take a look at bogus metrics being used to substantiate your behavior, which of course as humans is subject to vary every time we change our minds, so a lot of flawed data reported about you here with their scores. The pharmacies have to make their numbers, not so much worry about how they educate you, just make those numbers using this flawed software.
Patients Who Pay “Cash” When Filling Prescriptions Are Now Called “Outliers, Pharmacists Required to Fix Outliers as They Show Up As Non Medication Adherence Compliant With 5 Star Systems Full of Flawed Data…
Again this is a calling for those in our Congress to stop and do something for the consumer, give us at least the opportunity to know who’s scoring and selling our data? We have to fight years ago to get that right for credit data and it’s gone way beyond that now with bringing in all other types of business entities with their secret scoring. I guess at some point when the data scientists correlate to the point to where their spurious numbers are too far even for the dupes of hazard society to believe, well who knows what we might see, attacks on data centers? So here’s my campaign that’s been out there for a few years now and I predicted this about 3 years ago too that data selling with skewing context about consumers would be the greatest attack on our liberties we’ve ever seen, all for the sake of corporations make money. Scroll down and watch the 4 videos in the footer for some additional information from folks smarter than me. BD
0 comments :
Post a Comment