Follow us

0
87

Features

Mobile Data Challenge 2012: Unlocking the secrets of smartphone data

Boc Ly Published by Boc Ly Wed, Jun 20

Mobile Data Challenge 2012: Unlocking the secrets of smartphone data

0
87

Boc Ly Published by Boc Ly Wed, Jun 20

Lights over Europe seen from Space

A smartphone can say many things about us, but what can be revealed through our data usage?

Surely, this is just a lot of technical information that holds little value?

A joint initiative between Nokia and the Idiap Research Institute in Switzerland has sought to find the answers to these questions.

From late 2009 to early 2011, 200 volunteers were each given smartphones with special data-collection software in the Lake Geneva region.

During this period, information such as GPS, social interaction logs, calls, messages and app usage were captured.

Mobile device as sensor

This gathered data was then released to the research community and has formed the basis of the Mobile Data Challenge 2012 (MDC).

What would the data show and what ideas could it generate?

“This data captured by mobile devices tells you a lot about the behaviour of the people and in a way what was not possible by other types of research, like from questionnaires,” says Juha K. Laurila of the Nokia Research Center in Lausanne.

“Behavioural science researchers can now analyse new types of findings, which previously haven’t been easy or possible to measure without having data that is generated by devices carried by the research subjects 24/7.”

Mobile Data Challenge 2012

Around 500 researchers from all over the world have taken part in the MDC 2012 and each project differed in its approach to the data, or looked at it from a different angle.

MDC organising committee

Juha K. Laurila (centre) and other members of the MDC organising committee

Some of the findings came up with simple conclusions that are easy to understand.

Such as when looking at the correlation between personality and smartphone data, one project found that extroverts are more likely to speak longer on incoming calls and are more likely to receive calls in return. It found also found that, perhaps not surprisingly, introverts are more likely to browse on the web.

Other projects, such as this one seen in the video below, from City University London, used visual analytics to demonstrate the usage of social networks spatially and through time.

Predictions

An important aspect of the MDC, says Juha, was using the data for predictive purposes.

“If the device is going to know in advance what the user is going to do next, then the phone can recommend something that you might need next or it can give you some options, which are most likely highly relevant for you,” he says.

This predictive capability was key to one of the winning MDC papers, Interdependence and Predictability of Human Mobility and Social Interactions, announced at the Pervasive Conference in Newcastle, UK, yesterday.

This paper, from the University of Birmingham, found that you could improve the predictive accuracy of where someone was going next if you also looked at the mobility information of his or her social group.

Other research papers and materials can also be found on the MDC website.

Privacy concerns

So it is clear that analysing smartphone data can highlight valuable information about how we live our lives.

With more six billion mobile subscriptions in 220 countries worldwide the amount of data gathered in the future could be frightening.

Men on their phones

Mobile devices are now woven intimately into the fabric of our lives and are embedded with more and more sensors. They can monitor us while we exercise, track us while we drive and pinpoint our precise location at any given minute.

Of course, for the purposes of the MDC, all the data gathered came from the 200 volunteers in Switzerland. However, on the issue of privacy, Juha says:

“First of all the user needs to be in control and in control of the data that has been captured. The user needs to understand what has been collected, what it is for and also that there is a control mechanism for the user to opt out or eliminate the data.”

Furthermore, he says that before sharing the data with the researchers, they removed the Personally Identifiable Information (PII) from the data.

Data Community

The topics and contributions from the MDC ranged from the deeply scientific to some more light-hearted analysis, but many of them can be utilised by the mobile industry or by developers.

These efforts are just the beginnings of a global research community looking at smartphone data usage.

Says Juha:

“All in all, from the mobile industry and user point of view, we can better understand mobile usage and we can make better experiences for our users by making better devices, applications and services become more adaptive to the user situation.” 

image credits: woodleywonderworks and Menno van der Horst

comments powered by Disqus