Smartphone data reveal
Irma: back in control of your personal data
Figure 1: Data from 717,527 people’s smartphones from over 68 million days of operation show global variation in physical activity. Figure 2: Exercise disparity is linked to obesity and widening gender disparities in physical activity. Figure 3: Aspects of the built environment, such as walkability, can help to reduce gender disparities in operation and overall inequality.
Tables of Contents
Tables 1-3 are included in this file. The dataset statistics for the 46 countries with more than 1000 subjects are summarized in Table 1. Table 2 lists the cities in the United States in order of their walk ratings. Table 3 depicts three cities in the United States that are similar geographically. Table 4 indicates the number of subjects in the walkability study for each city and community. (474 kb PDF) Slides for PowerPoint
By leaving a comment, you agree to follow our Terms of Service and Community Guidelines. If you come across something that is abusive or does not follow our terms and conditions, please report it as unacceptable.
Find out who’s tracking you through your phone
Every minute of the day, hundreds of companies — essentially unregulated and unmonitored — log the movements of tens of millions of people using cell phones and store the data in massive data files all over the world. One such file, by far the largest and most confidential ever checked by journalists, was obtained by the Times Privacy Project. It contains more than 50 billion location pings from more than 12 million Americans’ phones as they traveled through major cities such as Washington, New York, San Francisco, and Los Angeles.
Each item in this file reflects the exact location of a single smartphone over the course of several months in 2016 and 2017. The information was given to Times Opinion by anonymous sources who requested anonymity because they were not allowed to share it and could face serious consequences if they did. The information’s sources said they were concerned about how it could be misused and felt compelled to warn the public and lawmakers.
How cell phones reveal your location – computerphile
As seen from the Washington Avenue overpass at 10 a.m. Sunday, Interstate 295 in Portland is eerily silent. Movement has decreased in southern Maine’s heavily populated urban areas, although decreases in some rural areas have been less pronounced. Staff Photographer Ben McCanna
As company work-from-home practices transitioned into school and business closures, and eventually a statewide stay-at-home order from Gov. Janet Mills, mobility in Maine has slowly declined over the past five weeks, according to state traffic data and publicly accessible smartphone metadata.
According to traffic data from the Maine Department of Transportation and smartphone tracking data from Unacast, a New York-based “human mobility data company,” changes in mobility vary greatly by area. Movement has collapsed in southern Maine’s urban and more heavily populated areas, although decreases in some rural areas have been less pronounced.
Kathryn Ballingall, a transportation researcher at the University of Maine’s Margaret Chase Policy Center, said, “The data needs to be taken with a massive grain of salt.” “It’s possible that there are statistical flaws here that are unconsciously skewed towards rural areas.”
How a smartphone knows up from down (accelerometer
The sensor that controls this work, known as a triaxial accelerometer, has been discovered by Gary M. Weiss, Ph.D., to be able to ascertain precise details about who is keeping the phone and what task he or she is conducting.
The principle is to use smartphone data to make observations about who a person is and what he or she is doing at any given time—whether walking, running, jumping, climbing stairs, or sitting.
“Assuming we’re not a hacker trying to sell this stuff, why would we want to do this?” We want to use data to the fullest extent possible as data miners. “We want to learn a lot about you and use what we learn in an ethical way,” he said.
This data may be used to send a call to voicemail if the phone detects that a person is running, turn to inspiring music if a jogger is slowing down (for example, the Rocky theme), or call 911 if an elderly user falls.
Weiss discovered that his team’s study correctly predicted what a person was doing 72 percent of the time based on a survey of 36 participants. When a model was personalized for a single user, it was nearly 100 percent accurate.