Software used to target ads for rental cars has been successfully applied to keeping British youngsters in education or employment after leaving school.
The predictive risk modelling software from IBM was turned to an unusual use by the Kent-based Medway Youth Trust after an employee had a brainwave.
Aware that it is easier to help kids before they get into trouble rather than when they're unemployed, unqualified and dabbling in petty crime, Data Quality Manager Gary Seaman decided to apply customer behaviour algorithms to information about young people to predict who was likely to wind up in trouble so that the youth services could reach them before they did.
And they decided to use software bought by companies like Avis and life insurers. The result is the Hidden Patterns social enterprise that aims to rescue kids with data. IBM donate the charity their predictive behaviour software for free and the trial has been running since February.
Using data to predict who's going to get into trouble
Medway CEO Graham Clewes told us that out of the 732 Year 11 students identified by the software in February, 648 were currently in some kind of further education or job: an outcome which is more positive than expected and has saved the 16-17 year olds who were all at risk of dropping out of the system from becoming NEETs, the hard to reach "Not in Education, Employment or Training" group. NEETs ends up costing the government hundreds of thousands of pounds in benefits, healthcare and other problems.
"It's much easier to get to a young person if they're still in learning than if they've been out for several months," says Clewes.
The software has saved weeks of staff time as research tasks that could have spanned a fortnight were whittled down to a matter of hours when automated by a computer. It also helps the Medway Trust access and correlate info that would otherwise have been lost.
The stats that show you're on the wrong path
The IBM software is particularly powerful because it will search through text as well as statistics. This means it brings in information that other data scrapers miss and provides a more complete picture than previously possible.
Information scraped by Medway software includes: CVs, medical records, school reports, write-ups of interviews with youth workers, social care reports and statements by the young person.
That info is combined with date of birth, ethnicity and reports on the young person's family situation.
"We looked at a lot of technology companies," says Clewes. "The key thing is that the IBM software analyses text data ... it meant that we were able to draw out what patterns might be hidden."
For example, if a young person is getting alcohol counselling through their local church not through the NHS a straightforward trawl of stats wouldn't find that info, but if the issue has come up in a conversation with a youth worker, then it will be noted by the Hidden Pattern software.
Clewes stressed that all teens sign consent forms before their data gets used. It is kept confidential and teens can opt for their parents not to see the reports.
The software that can predict insurance sales - and teenagers
The kid-saving software is from the IBM SPSS Predictive analytics suite acquired by IBM when they bought out SPSS in October 2009. It analyses data to predict behaviour.
IBM explain that it uses:
Advanced mathematical and statistical expertise to extract predictive knowledge that when deployed into existing processes makes them adaptive to improve outcomes.
The modeller is the engine of the programme:
IBM SPSS Modeler enables you to discover hidden relationships in both structured and unstructured (text) data – and anticipate the outcomes of future interactions.
Medway Youth Trust has now set up a social enterprise which aims to help other charities, local authorities and central government organisations develop similar models for predicting and reducing NEET status among young people. Over 30 local authorities have expressed interest and Medway are in discussion with the department of Education. ®