Slashdot Mirror
Cell Phone Data Predicts Movement Patterns
azoblue writes "In a study published in Science, researchers examined customer location data culled from cellular service providers. By looking at how customers moved around, the authors of the study found that it may be possible to predict human movement patterns and location up to 93 percent of the time."
Hopefully they find a way to program those (!@#) traffic lights a little better with this!
Seeing how 66.67% of the time I am either sleeping at home or at work it shouldn't be too hard to fill the other 27% with commute/grocery shopping.
"All tyranny needs to gain a foothold is for people of good conscience to remain silent." [Thomas Jefferson]
JUST 93%?
this is why we don't want them knowing were we are... unless we want them to know were we are by say making an emergency call...
Donald 'Duck' Dunn: We had a band powerful enough to turn goat piss into gasoline.
Imagine that. If you study someone's daily routine you can "predict" where they will go. Call me shocked.
Is this really a surprise to anyone? I'd wager the day for the vast majority of people goes something like "wake up, work/school, home, sleep", with the removal of work on Saturdays and Sundays and the possible addition of church or something on Sundays. It's not really that hard to predict something that consistent.
While not to the exactness of this study, this has been done before in May 2009 ( http://www.pbs.org/newshour/updates/science/jan-june09/celldata_05-15.html ). From the article:
analyzed six months of anonymous cell phone records from more than 100,000 people in a European country, obtained from a European cell phone provider. Those cell phone records gave an approximation of each person's location at the time of each call, because cell phone calls are routed through the nearest cell tower. He and his colleagues found that people tend not to stray far -- almost three quarters of the people stayed mainly within about a 20-mile circle for the entire six months, and nearly half the people rarely strayed outside a six-mile circle. They also tended to go back and forth regularly between only a few locations, such as home and work.
And another attempt on the same idea was done by MIT in July 2005 ( http://yro.slashdot.org/article.pl?sid=05/07/25/1751234 ). Difference here was that the percentage was 85%. Not the 93% declared now. From the Wired article:
Eagle's Reality Mining project logged 350,000 hours of data over nine months about the location, proximity, activity and communication of volunteers, and was quickly able to guess whether two people were friends or just co-workers.... Given enough data, Eagle's algorithms were able to predict what people -- especially professors and Media Lab employees -- would do next and be right up to 85 percent of the time.... Eagle used Bluetooth-enabled Nokia 6600 smartphones running custom programs that logged cell-tower information to record the phones' locations. Every five minutes, the phones also scanned the immediate vicinity for other participating phones. Using data gleaned from cell-phone towers and calling information, the system is able to predict, for example, whether someone will go out for the evening based on the volume of calls they made to friends.
Attention... all grammer nazi"s! Is they're anything; wrong with: my post,
So you're saying that analyzing movement patterns allows you to predict movement patterns. Would you like to guess the color of my red car?
Am I the only one concerned at random researchers keeping track of where I am, where I went and where I'll probably go? I'm not ok with some people *I know* knowing my schedule, let alone random people.
I see no valid reasoning for this study to intrude in privacy like this, since from the get-go it didn't aspire to answer any meaningful question: proving that you're able to ascertain someone's schedule from their phone calls seems like a very sordid thing to prove.
'Never' falls within 'up to 93% of the time' n'est ce-pas?
Sounds like a High School science project from a lame student; how are these results relevant or interesting at all?
Dear
BRB...this girl I used to date will be at Subway in five minutes. Time to casually bump into her.
It seems like you haven't used the WC for any extended period in two days.
Your local grocery store now has wholebran cereal on sale.
Just mention this ad and we will knock an extra 2 dollars of the already low price.
Coupon code: knst1patd
This seems to be the upper bound of predictability by computers ; in other domains of artificial intelligence, such as automatic translation or speech recognition, automated statistical analysis from corpuses seems to perform better than manual encoding of rules, but ends up at maybe 90% efficiency. The rest is too random to be predicted, and it could be the part of poetry, art or intelligence in our lives.
Google passes Turing test : see my journal
More and more we see stories like this, if you can connect the dots. this is a NASA coverup for Planet X, pure and simple. Fireballs almost every day or at least reported on every week, fallin' out of the sky. it'll be happening more and more, then NASA will fall silent. They do not tell the truth of what's out there, if you don't believe this: your programmed. Crow.
This is a really old arstechnica article. Wasn't this article from last week?
Arstechnica is a great site, although they do tend to get carried away with their stories. I really read arstechnica when I'm in a boring class and need to pass time because their articles are so ridiculously long.
I predict you're in a non-mathematical field, perhaps banking.
I can also predict that he's not an experimental physicist and has no access to instant-teleport technology either. ...and he might also be a little bit slow when he walks between his house, work place, shop and car (or other vehicle he's using when commuting).
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
it may be possible to predict human movement patterns and location up to 93 percent of the time.
The remaining 7% are deviant, they're probably onto some terrorist task of some sort and should be Guantanamoed.
I don't have a cellphone. How predictable are humans without cellphones? I might do something unexpected, I hope I don't get arrested for being unable to be profiled! Oh wait, it's not 1984 yet, that's just the future phobia...
(btw, my captcha was 'escapes')
What about the user who carries the phone from home to work and back, and never carries it anywhere else? Can they predict that movement? How about those of us who don't use a cellphone to begin with? I'm a web designer and a landline serves my needs just fine, thankyou. Track me, I wish you luck ;)
Sheesh what a useless study,
98% of people will go to work in the morning and come home from work in the afternoon. Assuming there are no detours they will take the same route every day. When they aren't working 72% of people will go to the grocery store once per week, and 27% of people will make up statistics to seem important.
Explaining the joke ruins the joke, but here I go.
Yes, 66.67% and 27% add up to 93.67% and not 100%.
You joked that this is probably due because the parent fails at math and can't come up with the correct number.
I joke that in fact, the total is correct. The missing 6.33% are the time spent walking between the various places mentioned :
He doesn't have access to an instant teleport device, therefore he can't teleport magically himself from the store into the car and his groceries bags into the trunk. He has to walk, and load the bags himself into the car. Albeit he probably walks rather slowly if the walking makes up the whole missing 6.33%.
So there go the missing 6.33% (well, that and time spent in the bathroom).
"predict human movement patterns and location up to 93 percent of the time."
I am either at work or at home, thank you very much for the study Dr. Obvious.
I am Bennett Haselton! I am Bennett Haselton!
All this Cell phone data should be in the public domain. There should be no privacy. Then my theft ring can predict when all your family are going to be out of the house. Cross reference this with your Twitters and annual vacation travel and we can really clean you out. Or perhaps the police just need to see if you have tags on all your pets. Or a quick look at your TVs to make sure you are paying that California Plasma tax. After they see what they need, then they can get the warrant for a REAL search. Maybe you have a cute daughter, I could show up at her favorite shop. Cross reference with her Rx and know when she is on the peak of her cycle. Look at credit card charges, give her some of her favorite treats. Perhaps know when she rides her bicycle home from school and goes through that dark patch of woods. Its a brave new world and Uncle Sam wants all your info available to anyone with hacker skills.
There is a new science evolving around understanding objectively humans and their behaviour.
As the century of the Self was driven by Freud this century will be driven by quantifying the self and even the culture.
Albert-Laszlo Barabasi the leader of this domain says we should get used with the idea that we are "dreaming robots on autopilot" nothing more.
I learned this four years ago on Law and Order: Special Victims Unit.
Those T.A.R.U. guys are some smart coppers.
Benson and Stabler know where every perv in New York is located.
Joe Dougherty, Florida, USA
The words I thought I brought, I left behind. So, never mind.
I'm at work or at home at least %93 percent of the time.
People are creatures of habit thus predictable most of the time.
So where is my cut of the grant money?
---- Booth was a patriot ----
You’re lucky when you get it down to 500 m.
Then again, when you’re a A-GPS server provider... ;)
Any sufficiently advanced intelligence is indistinguishable from stupidity.
I don't mean old as in "last week", this came up months ago, if not even pre 2009!
Not with cell phones, but some study was claiming to predict human location based on a study of your previous location, to a high degree of accuracy.
Now if I could just find the original link... I'm sure it was on slashdot...
Download them while you can and spread the word
http://www.public.asu.edu/~huanliu/sbp08/program.html
Papers:
Integrating Multi-Agent Technology with Cognitive Modeling to Develop an Insurgency Information Framework (IIF)
LeeRoy Bronner and Akeila Richards (Slides)
Stochastic Opponent Modeling Agents: A Case Study with Hezbollah
Aaron Mannes, Mary Michael, Amy Pate, Amy Sliva, V.S. Subrahmanian and Jonathan Wilkenfeld (Slides)
An approach to modeling group behaviors and beliefs in conflict situations
Norman Geddes and Michele Atkinson (Slides)
Computational Models of Multi-national Organizations
Alexander Levis, Smriti K. Kansal, A. Erkin Olmez and Ashraf M. AbuSharekh (Slides)
Clustering of Trajectory Data obtained from Soccer Game Record -A First Step to Behavioral Modeling
Shoji Hirano and Shusaku Tsumoto (Slides)
Inferring Social Network Structure using Mobile Phone Data
Nathan Eagle, Alex (Sandy) Pentland, and David Lazer (Slides)
Human Behavioral Modeling Using Fuzzy and Dempster-Shafer Theory
Ronald Yager (Slides)
Behavior Profiling for Computer Security Applications
David Robinson, George Cybenko, and Vincent Berk (Slides)
Mining for Social Process Signatures in Intelligence Data Streams
Robert Savell and George Cybenko (Slides)
An Ant Colony Optimization Approach to Expert Identification in Social Networks
Muhammad Ahmad and Jaideep Srivastava (Slides)
Modeling and Supporting Common Ground in Geo-Collaboration
Gregorio Convertino, Anna Wu, Craig H. Ganoe, Luke (Xiaolong) Zhang, and John M. Carroll (Slides)
Modeling Malaysian Public Opinion by Mining the Malaysian Blogosphere
Brian Ulicny (Slides)
Reading between the Lines: Human-centered Classification of Communication Patterns and Intentions
Daniela Stokar von Neuforn and Katrin Franke (Slides)
Automating Frame Analysis
Antonio Sanfilippo, Lyndsey Franklin, Stephen Tratz, Gary Danielson, Nicholas Mileson, Roderick Riensche and Liam McGrath (Slides)
Metagame Strategies of Nation-States, with Application to Cross-Strait Relations
Alex Chavez and Jun Zhang (Slides)
Using Topic Analysis to Compute Identity Group Attributes
Lashon Booker and Gary Strong
Posters:
Conceptualizing Trustworthiness Mechanisms for Countering Insider Threats
Shuyuan Mary Ho (Slides)
Particle Swarm Social Model for Group Social Learning in Adaptive Environment
Xiaohui Cui, Laura Pullum, Jim Treadwell, Robert Patton and Thomas Potok (Slides)
Social Network Analysis: Tasks and Tools
Steven Loscalzo and Lei Yu (Slides)
Behavioral Entropy of a Cellular Phone User
Ram Dantu, Santi Phithakkitnukoon, and Husain Husna (Slides)
Community Detection in a Large Real