Paedophiles who hide behind false online identities to groom children for sex could soon be flushed out by a computer programme.

Scientists have been working with police to develop a tool that can identify a computer user’s age and gender by analysing their language.

They hope the software will shorten the time-consuming process of examining logs of internet chat and other online material in the search for suspicious behaviour.

Professor Awais Rashid, of Lancaster University, said an early working version of the device could be available by next summer.

He said: ‘‘It picks up on subtle clues that an ordinary person would not be able to pick up on. That is where there is an operational interest.

‘‘The software looks at a range of things, for example, the structure of sentences, the language that is being used and also things that indicate deception.’’ Researchers brought in 350 school pupils from the Queen Elizabeth School, in Kirkby Lonsdale, Cumbria, to help them fine-tune the software.

They discovered that four out of five children (18%) could not tell when they were talking to an adult posing as a child.

The computer software did significantly better by correctly working out who wrote web chat in 47 out of 50 cases.

Last week, postman Michael Williams, 28, admitted 27 specimen charges, including grooming children for sex online through Facebook, Bebo and MSN messenger.

He used at least eight fake identities to target young people he met on his round, doing school runs as a taxi driver and as the secretary of a football club.

In March, sex offender Peter Chapman, who used a fake Facebook profile to lure Darlington student Ashleigh Hall to her death, was jailed for a minimum of 35 years.

Professor Rashid said the software could eventually track paedophiles as they move around the internet after identifying their stylistic ‘‘footprint’’.

He said children rely on subject matter, use of slang and the answers to simple questions to establish who an individual is when online.

Professor Rashid said he is working with the Child Exploitation and Online Protection Agency (Ceop) and several regional forces, which he declined to name.

He added: ‘‘In our analysis, we found that four out of five children across the school got it wrong.

‘‘Interestingly, the strategies they use to detect who they are talking to are also the ones that lead them to make wrong decisions.

‘‘They rely on the subject matter, use of slang and even something as simple as whether the individual said he or she was an adult or a child.

‘‘This really highlights the need for a safety net of some sort.’’ Queen Elizabeth School deputy head teacher Alison Wilkinson said: ‘‘We are concerned about the dangers our pupils face when they’re online and have welcomed the opportunity to help the project at the same time as raising pupils’ awareness of the risks.

‘‘It has been chilling to watch them being taken in by adults masquerading as teenagers.’’ The research project involves experts at Lancaster, Swansea and Middlesex universities.