Getting More Value from Cell-Phone Data
Last month, attendees at the World Economic Forum in Davos, Switzerland, pulled out their smart phones. Each attendee watched as the gadget displayed an analysis of his or her mental state—relative to others in the room—gleaned from data gathered by the phone itself.
A special program, which they’d agreed to install earlier in the conference, mined accelerometer data to tell when the phone had been moved, counted time spent on e-mail, and noted how long they’d kept the phone unlocked, and where and when they connected to Wi-Fi. “We could point out ‘you’re not very active right now,’ or ‘you’re not paying attention to this lecture,’ or ‘you’re jet lagged,’” Alex “Sandy” Pentland, the professor of media arts and sciences at MIT who developed the program, told the audience.
In the view of researchers like Pentland, the proliferation of smart phones—and our attachment to them—presents a prime opportunity to measure unseen behaviors and social interactions and use new algorithms to mine the value of this data. He calls this “reality mining,” and says it can make companies more innovative and deliver new marketing insights.
Some initial applications of cell-phone data have already been commercialized. For example, Sense Networks, a startup cofounded by Pentland, uses the ever-growing streams of real-time location information from cell phones and navigation devices—combined with historical research on consumer behavior—to predict the purchasing intentions of individuals across cities. Data on locations visited can indicate whether a person is going out for drinks or spending the day car-shopping.
The Sense Networks algorithms also continually update their analysis of anonymous cell-phone data to predict for clients whether the foot traffic in their neighborhood is, say, currently made up of business travelers or young adults out on the town. It looks at what kinds of locations they are visiting—such as a conference hall, rock club, or a hotel—and what times they are visiting them.
But Pentland sees many more commercial possibilities for cell-phone data. In a 2009 campus experiment, Pentland and a grad student used text-messaging and call logs, e-mail activity, GPS, and Bluetooth data from student volunteers’ cell phones to track informal social influence. Mining cell-phone data—and then surveying the students about what they were doing to create models that can help interpret future cell-phone data sets—allowed Pentland to track how students influenced each other’s political opinions, eating habits, or even their illnesses.
Pentland is known for developing a gadget called “sociometer” in the late 1990s. That’s a device worn around the neck that measures unspoken communication: a Bluetooth radio detects other nearby sociometers; infrared sensors detect face-to-face conversations; and a microphone picks up changes in intonation that can reveal who is the most influential speaker in a conversation.
In one experiment, sociometers revealed that gossip during coffee breaks at a call center for a major bank actually functioned as an exchange of informal tips about how to handle calls better. Pentland said changing the break schedule to optimize socializing made workers more productive and saved the call center $15 million a year.
Now “phones are sociometers, quite explicitly,” said Pentland.
Nicholas Christakis, a Harvard sociologist and physician, says that cell phones are one of a few sources—others include e-mail and institutional and research databases—that can provide rich streams of data to uncover new social interactions that can be of value to business. “All of these projects are interested in using new tools to understand fundamental problems,” Christakis said.
Of course, most data generated by cell phones is private. To develop more applications relevant to business, Pentland says, this issue will have to be addressed. Even in his small experiment at the World Economic Forum, individuals had to share their data for the program to make meaningful comparisons.
Pentland says he’s working on a “trust framework for privacy” with the Berkman Center for Internet & Society at Harvard University to give individuals control over data while participating in new research.
The framework is still in development, but Pentland’s goal is to create a tiered system where anyone who wanted to offer data—and perhaps in return use an application that leverages such data—could choose a level of privacy: a “bronze” tier would restrict data sent by the phone, and a “silver” or “gold” tier would send much more data.
Lauren Cox is a reporter for Technology Review.
Geoffrey Hinton tells us why he’s now scared of the tech he helped build
“I have suddenly switched my views on whether these things are going to be more intelligent than us.”
ChatGPT is going to change education, not destroy it
The narrative around cheating students doesn’t tell the whole story. Meet the teachers who think generative AI could actually make learning better.
Meet the people who use Notion to plan their whole lives
The workplace tool’s appeal extends far beyond organizing work projects. Many users find it’s just as useful for managing their free time.
Learning to code isn’t enough
Historically, learn-to-code efforts have provided opportunities for the few, but new efforts are aiming to be inclusive.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.