According to researchers, the amount of time a person spends on different smartphone apps is enough to distinguish them from a broader group in more than one-third of cases, with serious consequences for security and privacy.
780 people’s smartphone data was analysed by researchers from the Universities of Lancaster and Bath.
They used statistical models to analyse data from 4,680 days of app usage. Each of these days was paired to one of the 780 users, allowing the models to learn about their daily app usage patterns.
The researchers then examined whether models could identify a person using only a single day of anonymous smartphone behaviour that had not yet been paired with a user.
David Ellis from the University of Bath, said “Our models, which were trained on only six days of app usage data per person, could identify the correct person from a day of anonymous data one third of the time,”
While this may not appear to be much, when the models predicted who the data belonged to, they could also produce a list of the most to least likely candidates.
According to research published in the journal Psychological Science, it was feasible to see the top 10 most likely individuals to whom a given day of data belonged.
The correct user would be in the top ten most likely possibilities about 75% of the time.
Professor Paul Taylor from Lancaster University, said “In practical terms, a law enforcement investigation seeking to identify a criminal’s new phone from knowledge of their historic phone use could reduce a candidate pool of approximately 1,000 phones to 10 phones, with a 25 per cent risk of missing them,”
As a result, the researchers warn that software with access to a smartphone’s standard activity logging could render a reasonable prediction about a user’s identity even if the user was signed out of their account.
With monitoring of conversations or behaviours within apps themselves, identification is possible.
Heather Shaw from Lancaster University, said “We found that people exhibited consistent patterns in their application usage behaviours on a day-to-day basis, such as using Facebook the most and the calculator app the least. In support of this, we also showed that two days of smartphone data from the same person exhibited greater similarity in app usage patterns than two days of data from different people,”
Hence, it’s critical to recognise that app usage data, which is frequently collected automatically by a smartphone, has the ability to reveal a person’s identify.
According to the researchers, while this type of data provides new opportunities for law enforcement, it also poses a risk to privacy if it is misused.