This article is more than 1 year old
Smart phone predicts owner's behaviour
I'm sorry, Dave, I can't let you have another beer
US boffins are developing mobile phones which learn user's daily habits so that they can become "mobile digital secretaries".
Going beyond the calendar feature common in many current mobiles, the "smarter smartphone" learns about people's preferences by logging calls and noting when application like cameras are used. Location-based functions allow the phone to keep record where you work and socialise. The phone also makes note of Bluetooth pairing bonds, in theory allowing it to build a profile of who you socialise with. This information would be sent to a server which processes data and returns suggestions or reminders.
Beyond predictive texting the phone is touted as a device that predicts what you will do. The New Scientist reports possible applications include reminding you not to drink too much the night before an important presentation. Some people might balk as the idea of being monitored - and nagged - by their personal technology. But US scientists reckon they've hit on a winner.
The technology is the brainchild of Nathan Eagle and Sandy Pentland of the Massachusetts Institute of Technology. The system is based on mobile messaging software called Context, written by developers at the University of Helsinki and Helsinki Institute of Technology led by Mika Raento. The software build a profile of user's routine by asking them what they're up to when they come into range of a new mobile mast.
The New Scientist reports that the software has been installed on 100 of Nokia 6600 smartphones in a trial involving MIT students. Data is downloaded onto a server at MIT and processed using pattern recognition software. Boffins reckon the phone can help students work out how long they have spent partying and working in a week or how long it is since they last saw a friend. It might even be able to work out the strength of a friendship.
Results from the trial could be useful to researchers investigating how social networks build as well as technologists, New Scientist reports. ®