Slashdot Mirror
London's Crime Hot Spots Predicted Using Mobile Phone Data
KentuckyFC (1144503) writes A growing number of police forces around the world are using data on past crimes to predict the likelihood of crimes in the future. These predictions can be made more accurate by combining crime data with local demographic data about the local population. However, this data is time consuming and expensive to collect and so only updated rarely. Now a team of data experts have shown how combing crime data with data collected from mobile phones can make the prediction of future crimes even more accurate. The team used an anonymised dataset of O2 mobile phone users in the London metropolitan area during December 2012 and January 2013. They then used a small portion of the data to train a machine learning algorithm to find correlations between this and local crime statistics in the same period. Finally, they used the trained algorithm to predict future crime rates in the same areas. Without the mobile phone data, the predictions have an accuracy of 62 per cent. But the phone data increases this accuracy significantly to almost 70 per cent. What's more, the data is cheap to collect and can be gathered in more or less real time. Whether the general population would want their data used in this way is less clear but either way Minority Report-style policing is looking less far-fetched than when the film appeared in 2002.
Crime reduction is certainly a worthy reward, but as the article says, lots of people might not be too happy with having their information shared this way. Let's hope it is truly anonymous (which I doubt) and see how it goes.
of spamming with unreadable links.
More phones in an area = more people. More people = more crime.
"Don't belong. Never join. Think for yourself. Peace." V.Stone, Microsoft Corporation
If your mobile detects you are going into an area where it is likely to be stolen, does it start to vibrate nervously, or bleep as warning? Maybe we just need an app that automatically sounds an alarm "help I'm going to be stolen any minute now", presumably followed by "no no aaaaargh" and "that's another fine mess you've gotten me into" before self-destructing.
Slight danger is the user putting the sensitivity too high, but that's all good for people selling replacements...
Right in Slashdot !
"Let's hope" increasingly is not good enough. There is a point where just going forward without applying what we learned or even learning from what happened before in similar situations (your "hope" here) becomes criminal negligence. If we're not past that point for everybody yet, and apparently not past it for those who should be paying attention to this sort of thing, we're certainly past it for those who do pay attention to what happens in this space.
This has nothing to do with caucasions.
fuck them. Almost 2/3 prediction from existing crime stats. Gee I know a lot of cops aren't the brightest but really? Thats not enough of a leg up?
The "machine learning algorithm" is a euphemism for three hairless teenagers floating in pools of milk.
Watch out for the spiders.
I'm sorry, but your opinion seems to be wrong.
62% to 70% isn't exactly groundbreaking for something that varies greatly. This increase looks suspiciously like selecting results for passing a statistical test instead of using a statistical test to verify the significance of a given result. Relevant xkcd: Significant.
Also, there is no such thing as anonymised phone data.
This just in: If you track the location of criminal's cell phones you can predict areas at higher risk for crime.
Expounding on your statistics point as I agree that there is no significant increase in accuracy, notice the key phrase in the article.
The team used an anonymised dataset of O2 mobile phone users in the London metropolitan area during December 2012 and January 2013. They then used a small portion of the data to train a machine learning algorithm to find correlations between this and local crime statistics in the same period.
In other words, they took everything they gathered and pulled a subset that matched criteria that would back the claim that they could detect future crimes.
Computers can surely show what law enforcement already knows. E.G. That area is a known crime area. Computers don't make tea leaf reading possible, which is the claim that both Governments and Tech companies peddling software claim. Even worse, this type of technology does absolutely nothing to address the problems that actually cause most criminal activities. It exacerbates those problems because the economy this generates does not transfer down to desperate and impoverished people.
-The wise argue that there are few absolutes, the fool argues that there are no probabilities.
Any article citing statistics is invalid when they don't understand the difference between percent and per cent. Getting 62 things right per US penny is a VERY cost effective system, probably regardless of what information we want to get right.
Unfortunately, all this says is that if we place our population under total surveillance with trackers, we can increase anticipation of crime by 8% (accuracy of 62 to ALMOST 70%). This says nothing about preventing those crimes or what type of crimes it prevents. Actually if you read the article, it only increased accuracy to 68%, so a 6% gain. Way to glorify the stats in the media. They should have said "just over 60% to almost 70%". This would have made this 6% increase look like a 10% increase.
Big Brother is watching you. Again. Even when they say they're not.
(((dB)))
Are we in the future yet?
... in London... and in which parts...
But we're not allowed to say...
In what way does it more closely resemble "selecting results" than real results?
If I determine that this area is more likely to have a crime and increase police presence, then the crime doesn't happen because there's too much "heat" then haven't I skewed my results?
Or do you intend to have the cops lay low so they can "catch them in the act" or at least catch them quicker "after the fact"?
I refuse to sign
This is not like Minority Report at all. It predicts which locations at which times have a higher probability of a crime committing. It does not predict the particular crime, transgressor, or victim. It won't actually stop any crime from happening. The best it can do is allow a police force to more intelligently deploy their forces. They will be more able to rapidly respond to crimes after they happen, since statistically, they will more often have officers already dispatched to the nearby crime area.
"Love heals scars love left." -- Henry Rollins
"significantly"
I do not think that word means what you think it means.
"In other words, they found where the poor people live by looking at phone data."
No they found out that crime happens where the criminals are.
Groundbreaking.
New Dartmouth smartphone app reveals users' mental health, performance, behavior
Dartmouth researchers and their colleagues have built the first smartphone app that automatically reveals students' mental health, academic performance and behavioral trends. In other words, your smartphone knows your state of mind -- even if you don't -- and how that affects you.
The StudentLife app, which compares students' happiness, stress, depression and loneliness to their academic performance, also may be used in the general population – for example, to monitor mental health, trigger intervention and improve productivity in workplace employees.
"The StudentLife app is able to continuously make mental health assessment 24/7, opening the way for a new form of assessment," says computer science Professor Andrew Campbell, the study's senior author. "This is a very important and exciting breakthrough."
The researchers presented their findings on Wednesday at the ACM International Joint Conference on Pervasive and Ubiquitous Computing. The paper has been nominated for best paper at UbiComp, the top conference mobile computing. A PDF of the paper and a summary of the findings are available on request. They also released an anonymized version of the dataset in the hope that other social and behavioral scientists will use it in further studies.
The researchers built an Android app that monitored readings from smartphone sensors carried by 48 Dartmouth students during a 10-week term to assess their mental health (depression, loneliness, stress), academic performance (grades across all their classes, term GPA and cumulative GPA) and behavioral trends (how stress, sleep, visits to the gym, etc., change in response to college workload -- assignments, midterms, finals -- as the term progresses).
They used computational method and machine learning algorithms on the phone to assess sensor data and make higher level inferences (i.e., sleep, sociability, activity, etc.) The app that ran on students phones automatically measured the following behaviors 24/7 without any user interaction: sleep duration, the number and duration of conversations per day, physical activity (walking, sitting, running, standing), where they were located and how long they stayed there (i.e., dorm, class, party, gym), stress level, how good they felt about themselves, eating habits and more. The researchers used a number of well known pre- and post-mental health surveys and spring and cumulative GPAs for evaluation of mental health and academic performance, respectively.
The results show that passive and automatic sensor data from the Android phones significantly correlated with the students' mental health and their academic performance over the term.
Some specific findings: Students who sleep more or have more conversations are less likely to be depressed; students who are more physically active are less likely to feel lonely; students who are around other students are less likely to be depressed. Also, surprisingly, there was no correlation between students' academic performance and their class attendance; students who are more social (had more conversations) have a better GPA; students who have higher GPAs tend to be less physically active, have lower indoor mobility at night and are around more people.
The results open the door to the following breakthroughs for the first time:
your phone automatically knows if you are depressed, stressed or lonely
the phone sensor data can predict student GPA
coupled with intervention software, students can track their mental health and academic performance indicators with the goal of improving both
the app (and its methods) are applicable to non-student groups, such as workplace employees, with the goal of improving productivity or radically reducing stress -- your phone will know how productive you are on a daily basis.
"Under similar condit
TL;DR
Ummm, re " ... this [crime] data is time consuming and expensive to collect and so only updated rarely....":
Wrong! Not at all so. Most CAD's - Computer-Aided-Dispatch - including our own free, Open Source contribution at www.ticketscad.org - do exactly that. It's inherent to the task of dispatch management. AS
We know that people that commit crimes are much more often from certain social and cultural backgrounds. There are untold numbers of "anecdotal evidence" around, but we don't want that to be true. So we tell ourselves white lies, blame victims, discount hundreds of incidents as "anecdotal evidence", pinpoint the few cases outside the norm and fabricate elaborate excuses about why such and such were practically forced to commit crime. We are constantly telling ourselves how we are to blame for not paying enough welfare, not enough education, not giving enough leeway while conveniently ignoring millions of people of other social and cultural backgrounds that simply don't commit any more crime than everyone else, being good people despite being poor and uneducated.
Choices of cellphone contracts and handset make and models are similar along cultural and social bonds. An algorithm will never know about that but detect the significance.
But anyway, even among the groups with the highest part in crime, only a few select individuals are responsible for a large percentage of crime.
Algorithms will find that when IMSI xyz is in the general area, people will get robbed. It will also find that when expensive handsets with IMSI abc where in the area when a phone robbery happened, they will probably be around the next crime area as well, since the thief will either have it now or sold it to a pawn shop in the high crime area.
I don't buy it - an app that monitors every sensor, plus apparently monitoring abstract stuff like "stress level" somehow, 24/7?
Wouldn't that pretty much lock up and drain the battery of almost every phone on the market today? Hey, maybe that's how they determined stress level - using the accelerometer to determine how hard the student threw the phone against the wall when it froze up on them for the last time.
An enigma, wrapped in a riddle, shrouded in bacon and cheese