Slashdot Mirror
Innocent Until Predicted Guilty
theodp writes "Gizmodo has an angry piece on IBM helping Florida to predict how delinquent your child's going to be. The Florida Department of Juvenile Justice has decided to start using IBM predictive analytics software to help them determine which of the 85,000 kids who enter their system each year poses the biggest future threat. From IBM's sales pitch: 'Predictive analytics gives government organizations worldwide a highly-sophisticated and intelligent source to create safer communities by identifying, predicting, responding to and preventing criminal activities. It gives the criminal justice system the ability to draw upon the wealth of data available to detect patterns, make reliable projections and then take the appropriate action in real time to combat crime and protect citizens.'"
your child doesn't fall into the minority report.
It seems to me that if the government thinks it can predict these things and takes certain actions in prevention, it might actually cause the problem that is predicted, and thus validate the method.
With IBM software we can lock threads before flame wars start....
"Waste not one watt!" - CZ
Florida is insane, in the same way that senile demented octogenarians are insane. They never think past tomorrow, because they don't know if they're going to live until tomorrow. All that matters is today, the pudding, and Matlock.
You guys are being pretty hard on IBM. They're just providing computing and analytical power. You're acting like they collaborated with the Nazis or something.
The best part is, if the software doesn't currently consider you a "threat" we can always tweak it to push you over the threshold! Remember that come next election, or next time you purchase something we don't think you should, or even the next time you pass us and don't give us a compliment!
Do predictive analytics work for other demographics as well, e.g. middle aged white man from prominent Ivy League university running an energy company more likely to steal billions of dollars over young Latino kid living in downtown Miami?
I believe the best use of this technology is as a means for monitoring our government officials and representatives (starting with the folks thinking about using it here.) It is arguable that the harm done by the average juvenile delinquent pales in comparison to the social and economic harm done by politicians and lawless officials. We should be using predictive technology keep them in check, and ensure that liberty is being preserved for future generations...
This isn't even thought crime. That at least requires you to think about doing something "subversive". This merely requires an arbitrary set of parameters to flag you as potentially "subversive". No thought required.
Sure you can, the entire insurance industry is based on doing just that.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
New corporate slogan: "IBM an Integrated Best of Breed Final Solutions Provider"?
Okay, all I see here is a slippery slope argument. Juvenile delinquents who have been convicted of a crime are generally sentenced to probation, attendance in educational programs, counseling, etc. The only difference here is now they're using computer models to decide which programs are most appropriate for a given youth based upon the data they put in... instead of the court making the decision based upon a less complete set of data and a less methodical prediction of what would work best for that individual.
Now I'm not saying IBM's system works. It may or may not and that needs to be carefully studied. I have no problem, however, with computer models being used to determine which juvenile delinquents are most likely to benefit from specific programs and which are most in need of them when resources are limited. Appeals to various constitutional amendments are just empty rhetoric, given these kids have been convicted of a crime and this is part of their rehabilitation. In fact this whole article looks like an excuse for sensationalism and a reason to display cool graphics from "Minority Report". Lame Mr. Diaz.
The story doesn't give too much information, it's just a rant; I'm curious as to what "education" is going to take place; maybe it's a good thing.
I imagine that the software is a Bayesian filter that takes in lots of seemingly unrelated factors and combines them into a score. First, yeah, I get the obvious dystopian implications - I won't argue against the awful possibilities if it were widely deployed. That said, isn't it possible that it could genuinely help some kids? Suppose those factors like increased absences and a couple of minor contacts with police indicate that Johnny is extremely likely to drop out of school. Maybe that's a good hint that someone needs to talk to Johnny and see if something correctable is going on in his life.
Dewey, what part of this looks like authorities should be involved?
Sure you can. You just can't be particularly sure that it'll be accurate.
Luckily, the people most likely to be (questionably accurately) judged to be pre-crime risks are likely to be members of more or less unsympathetic and disliked groups, so the people who actually count won't much care whether you are accurate or not. If anything, the "good upstanding citizens" will howl with rage and demand that they tolerance for false positives be increased every time the blood-spattered story of a false negative hits the cable news...
Sandra Bullock and Tom Cruise are doing a movie together? I'm not sure my gag reflex is strong enough for that.
How we know is more important than what we know.
It all depends on what they do with this software. My reading of this article is that this is an expert system for judges who sentence juvenile offenders. Typically judges have discretion in sentencing youth. They research the background, number of offenses, etc of the offender and pick an appropriate program. However, they don't have all the data to make a better decision. Do Latino youth who committed a second non-violent offense respond better (get arrested less often in the future) to mental health treatments, mentoring programs, or incarceration?
This system seems to automate this process. So it is possible it will save money and produce better results than the current system, while still maintaining fairness. After all, if you have committed a crime, both the maximum and minimum penalties for what you did should be fair outcomes.
Actually, this is not an "octogenarian" problem.
It is a problem of having bloated and pretty much do-nothing administration.
Disclaimer: I live in this hell-hole jokingly called "God's Waiting Room".
Any reason why we are angry with this? The whole point appears to identify at-risk kids and make sure they get the support they need.
A: Kid is from a low income family
B: Kid lives in drug ridden neighborhood
C: Kid eats twice a day
D: Kid is in a single parent home
Kid is BLAH BLAH% likely to commit a violent crime.
A is 38% weighted
B: is 14% weighted
C: is 17% weighted
D: is 9% weighted
Per $ ROI indicates that an additional $4.22 spent weekly on school lunch program (C) will save $19.22 over 10 years in reduced criminal activity.
Blah blah blah...
Seems par for the course...
-=[ Who Is John Galt? ]=-
Of course, if you'd bothered to read TFA (and were able to ignore the author's histrionics), you'd realize that the idea is to use this technology to differentially sentence offenders based on the likelihood of recidivism. That is, juveniles who have already committed a crime.
The average child's behavior does have the potential to reliably predict future social and behavior patterns for the individual; however, there are outliers of varying types who would not be well served by this attempt at divining the 'future history' of individual human beings. Here are some of the types I have noticed.
1. Situational issues such as abuse at home that cause anger, frustration and inappropriate behavior at school. Children's brains are luckily plastic enough to rewire themselves when presented with a new environment that is far more nurturing, safe and empowering.
2. Schools/Neighborhoods that have been left to become warrens of crime will produce children that seek criminal behavior to 'fit in', even if they are articulate and attentive in class they may be encountering overwhelming peer pressure to conform to another set of behaviors outside the classroom or face ostracization.
3. Mentally ill children who go unmedicated can be hellions the days they don't take their meds and perfectly reasonable mature human beings when they do. The flip side of this, is dealing with the many popular NT rich kids whose parents have gotten them adderall prescriptions babbling in the back of the classroom and acting hyper aggressive on the playground.
4. Police provoked violence/crimes. I did some student teaching in a High School which shall remain unnamed and the MO of the high school police was to find the 'troublemakers' smoking cigarettes across from the school or in the alleyways surrounding and set up a cop car on one side of them and try to herd them towards it, if they ran they tried to take them down with tackling and submission holds. The kids got suspended and charged with resisting arrest at the very least some got thrown in Juvi all for smoking a cig and being confronted by a dickish bunch of cops.
An Education is the Font of All Liberty
This isn't news. SPSS already exists since 1968, and is now on version 18 of the software. IBM just bought the program in 2009. For those who never heard of the program: it's a souped up Excel with advanced statistics and datamining. Here at my work (public health department in Amsterdam) they use it a lot for scientific studies of health, surveys etc. In fact the use of SPSS in the field of research is so widespread for many years already it's strange they only replaced Excel with it now...
I'd guess Slashdot geeks would really like it since you can program some nice stuff in a pseudo SQL script language (I don't know the name of it), but if you've ever seen it you'll know that SYSMIS sorta means NULL.
Gizmodo links this technology to Minority Report, and certainly not without cause, but the movie that really ought to worry you here is Gattaca. What happens to kids this software flags with a high potential for future criminal activity? If companies start taking this data seriously, a lot of them won't be hiring these kids. And while it was genetics that was the profiling mechanism in Gattaca, considering we've already cracked the human genome, it can only be a matter of time before someone decides to take a similar piece of software and run it against someone's DNA.
Let's stop pretending that this is anything but a technological tool for doing what is already happening. Individuals are already differentially sentenced for all kinds of reasons, many of them terrible. Far better to use well-understood machine-learning/data-mining techniques instead of the discretion of individual judges and all its attendant biases.
N.B. This obviously has the potential for misuse (e.g. the first time some political hack suggests it is great for preemption.), but it is not a prima facie violation of individual's rights.
Eternal Vigilance, etc.
This was almost attempted in the early 70's. Look up the "Hartman Value Profile". It was shot down in flames, I guess the concept of Civil Rights has changed a bit since then . . .
Not just answers, the correct questions.
There is nothing about this that is 'pre-crime' or would have 'false positives'. This is about how to determine what to with people who have already entered the juvenile justice system (ie. post-crime), to try to rehab them. So, the question is, can analysis of risk factors for recidivism actual prevent recidivism? I don't know, but it seems silly to just dismiss it out of hand.
Now, consider IBM's press release, which seems to be the only news available on this subject, and is certainly the unlinked source of Gizmodo's fit. Previously, Florida State officials were using Excel macros to sort convicted juvenile offenders into different programs, and now they will use IBM's software to do it. Whether Florida's juvenile prosecutions or unjust or not, whether their programs are effective or not, has no bearing on IBM's part in this.
IBM has sold Florida some statistical analysis software, which they will (apparently) use to stick heavy offenders into more punitive detention programs, saving the spaces in more rehabilitative programs for newer offenders. You may think that that policy is ill-advised as well--but it is perfectly legal. At least the sorting will be (hopefully) less capricious than it was before. IBM is certainly not enabling Florida to enforce "pre-crimes" or anything of the sort. This is not even affecting the judicial sentences. Everyone being analyzed here has already been found guilty by a court.
Prior to predictive analytics, the organization used Excel for basic analysis on projections for the number of delinquency cases they would take in, which had limited functionality. They selected IBM SPSS predictive analytics due to the ease of use and the advanced analytic capabilities.
The organization will now utilize the new predictive analytics system as a component in many of the performance measurement analyses conducted and distributed to agency staff throughout the year. These reports assess the future of delinquency cases to evaluate what juvenile crime trends may look like in the immediate future. This information will help the organization to better plan and project staffing and other resource needs.
IBM recently also announced that the Ministry of Justice in the United Kingdom uses predictive analytics to assess the likelihood of prisoners reoffending upon their release to help improve public safety. With predictive technology from IBM, the Ministry of Justice is analyzing hidden trends and patterns within the data. IBM SPSS predictive analytics has helped identify whether offenders with specific problems such as drug and alcohol misuse are more likely to reoffend than other prisoners.
It sounds like the Ministry of Justice might have something a bit more Orwellian (notice "public safety") in mind, but that will be a story for another day. Now take a deep breath, and control yourselves next time Kdawson posts a link to an inflammatory and ill-informed opinion piece. A worthier title for this event might have been "IBM enables Florida Juvenile Detention System to Become Slightly Less Cruel and Arbitrary."
Did you mean, weak enough?
You know what? I can predict who will become involved in criminal activities. The answer to one simple question can peg it every time: ARE YOU HUMAN?
People are criminals. Republican, Democrat, bullshit. You ever speed? Criminal. You ever not come to a 100% complete stop before advancing past a stop sign? Criminal. Smoke weed? Criminal. Drink before the age of 21? Criminal.
Laws exist to make us criminals. We all are. We just get away with our crimes until we piss off the wrong people.
Take a step back here, guys. This system is NOT being implemented for children entering school, or anything like that. It is being used for children who are actually entering the juvenile justice system. Kids who are already very much in trouble. These kids are already being pigeonholed by caseworkers, DAs, judges, probation officers, and a host of outsiders of the system such as school officials.
They are actually trying to use some actual data to try to direct these kids within the justice system. That's not such a terrible thing - in fact, it's what people in the system who are trying to do the right thing are trying to do - get kids the help they need to save their future. Like anybody, I would hope that the software is not blindly relied upon, but the people in the system are still going to be there. It's pretty hard right now to fight for a kid to get the attention they need and be directed the way they should in that system. Because these are kids who are already being judged by the law, there's not really any worse situation they can get into. They can already have a judge or caseworker who arbitrarily hates them, or who wants to help them. The factors for re-offending are already being examined, it's just that now some software can spit out a report based on these factors.
There's not much to see here.
Let me get this off my chest first: Jesus Diaz (the Gizmodo writer) is an idiot of the most supreme caliber. His MO on the Gizmodo is to write ill-researched inflammatory articles. Over time I have figured that these must be written solely to stir up internet frenzy and increase page views. More power to him, but it automatically disadvantages his opinions for me.
Now that the ad-hominem is out of the way, let me get to the meat of it. The conclusions here are 100% wrong. What we do is provide Juvenile Justice departments (which is almost always Juvenile Probation) with tools, in the form of academically validated models, that help them determine which kids are at highest risk to re-offend. We're also able to determine, with a high degree of accuracy (thank you academia!) what the kids biggest needs areas are.
So how does all of this information actually get used? It turns out that it's used in amazingly great ways. It helps keep children placed in their own homes, not in residential treatment, juvenile hall, or the state's Youth Authority. We've had jurisdictions report out-of-home placements drop by 50% after implementing our tools. It also means that a Probation Officer can focus on kids that are at a high risk to re-offend, and have minimal contact with kids that are at a low risk to re-offend. As it turns out, the PHds that come up with these tools are able to determine that having lots of contact with the criminal justice system is bad for kids that are low risk - so it really helps to know the kids that minimal intervention is the best path for. Another benefit of this sort of classification scheme (which works just as well for adults) is that the officers (who are time constrained) are able to spend more time with their higher risk kids because they aren't spending as much time with their low risk kids. This probably seems obvious to most readers, but I'm surprised by the number of commenters that don't get that last point.
My final point is that these kids are already getting put into treatment programs, like anger management, or drug counseling, or teenage parent classes. That happens regardless of whether or not a jurisdiction uses software like ours. What this type of analytic software does is help take away the "gut instinct" part of program placement and give the officer a little more guidance into what programs will be most effective. If you can only send a kid to one program, why make it an anger management class when, after an assessment, you are able to determine that it's actually his drug use and poor school attendance that are his biggest risk factors?
So in the end, this isn't about pre-crime, or thought-crime, or any sort of Orwellian conspiracy. It is, quite literally, about helping place minors (and adults) that have already committed crimes against people or their community, into programs that have the statistically best chance of helping them not commit another crime. The best part is, the followup data from jurisdictions using this type of software suggests that it works, with fewer placements and less recidivism.
Oh, and Jesus Diaz is a idiot (man that feels good.)
Which is exactly what judges and parole officers do today using guesswork and Kentucky windage. I think it's hard to maintain that making that process more objective and automated is a bad thing.
Of course, there's potential for abuse through its misapplication to other areas.
This thing is a tool, just like most of the other ones that are out there. Your insurance (for pretty much anything, auto, life, house, etc.) use the same basic principle. You take the big factors, sort them out into groups, and then figure out the relative risk.
You're young, male, and single? Guess what, on average, you've got a better chance of being in a car accident. Live in Florida? On average, you've got a better chance of having a house being blown away by a hurricane. There's data support all of it, but you've got to remember that it's a mathematical predictor, which doesn't mean it's always true.
The real question is what the heck are they going to do after the risks are tabulated? As a post mentioned before, it seems that it may be used for sentencing guidelines, but I hope that's just a small part. If you can find out the groups of at-risk youth, and then do something (like spend money on programs and other things that will encourage those youth to not commit crime (I know, far cry of concept for our elected officials)), then the system is worth it. Prevention of crime doesn't always mean that the police are directly involved. Rather, they should be the last resort.
And as to "real time", it's about as real time (if they're using most modern analytical software) as the data they get in. Ask someone in insurance how quickly they get data in (bonus points if you can find out the differences in time for various coverages).
PS - I love the idea about using this for politicians, and comparing polls in their represented area to their voting history. I'd bet a lot of them wouldn't like that brought up.
And why should the government decide who goes to an specific prevention program or who doesn't based on what a computer says? The fact is that, even if the software was 99.99% accurate, there will be always an innocent person who will be F***ed. And that is exactly why we have something called due process and the presumption of innocence. That's why those things are not only in the United States Constitution, but in the Universal Declaration of Human Rights too.
Nice rant and straw man. They're not talking about throwing these kids in jail. Instead, targetting people at higher risk of following a criminal path, and giving them extra efforts to steer them from that path.
ISure. Some will argue that these juvenile delinquents were already convicted for other crimes, so hey, there's no harm. This software will help prevent further crimes. It will make all of us safer? But would it? Where's the guarantee of that? Why does the state have to assume that criminal behavior is a given?
Why would you naively assume that criminal behavior is no more likely among a population so identified than among the rest of the population? They're not just throwing allof the state's youth into the system to churn out who the "defectives" are. These are people who have a) already started down a criminal path or b) come from homes so screwed up that they had to be removed for their own safety. Statistically, both of these groups have a considerably reduced chance of a normal life. If there's a reliable means of figuring out which subset of those groups are more prone than others, and to take active steps to help them prevent it, why would you not do it? Why would you don the rose-colored glasses and pretend that they can just step out of Juvie and lead a fully normal life, when the odds are stacked against it?
Ok, I've got my flame-retardant suit on. Let's have it...
Lets actually think about this instead of just coming up with reasons to force it into are preconceived notions.
If there are a set of factors to indicate risk, then this can be a good thing.
Steps can be taken to changing factors and reduce risk.
For example: One factor could be determine by a lack of social skills. That child could be taught the normal social behaviours. Improving the child's life as an adult. No we are not talking about making everyone the same. Another example, there is a strong correlation between torturing animals and being a psychopath. If you find you child torturing animals, then you should get help for that child. Before you pound you meat stumps against the key board, I am not saying torturing animals causes it, only that it's a sign, so put away your misguided 'correlation is no causation' standard reply, it does not apply here.
I know minority report is a fun comparison, but the movies response to 'pre-crime' is nonsense. Plus that's not what we are talking about.
The Kruger Dunning explains most post on
"Mrs. King, we have determined that your son Martin is statistically likely to grow up to be a troublemaker. We advise you begin Prozac treatments to curb these tendencies and allow him to live a normal life. Now please move to be back of the bus."