'Anti-grooming' software can detect paedophiles

Click to follow
The Independent Online

Scientists have developed "anti-grooming" software which can tell young people if they are chatting to an adult posing as a child on the internet.

Trials among pupils at a secondary school have shown it to be remarkably effective in helping them spot other internet user's real ages.

The program, which has been developed by Lancaster University, can work out a person's age and gender by using language analysis techniques, detecting words which are often used by people of a particular age group. Computer experts believe it could be a significant breakthrough in helping to catch paedophiles.

At first, the 350 young people who took part in the trial – from the Queen Elizabeth School in Kirkby Lonsdale, Cumbria – were unwittingly exposed to an experiment to find out if they could tell when they were chatting to adults posing as children. Only 18 per cent correctly guessed whether they were talking to an adult or a child.

Approximately four out of five children thought they were chatting to a teenager when it was, in fact, an adult. Youngsters of all age groups at the school – even 17-year-olds – failed to spot the impostors. Girls were more adept than boys at guessing ages correctly, with a 22 per cent success rate compared to the boys' 16 per cent.

However, when the anti-grooming software was installed on their computers, it correctly worked out whether it was an adult or a child using a chatroom in 47 cases out of 50 – including when an adult was pretending to be a child.

"Paedophiles often pose as children online and our research indicates that children don't find it easy to spot an adult pretending to be a child," said Professor Awais Rashid, from Lancaster University's computing department.

"We hope to develop an automated system which can pick up on quirks of language particular to a certain age group. These language patterns can help us to expose adults who seek to groom children online.

"The software looks at a range of things, for example, the structure of sentences, the language which is being used and also things which indicate deception."

He added that teenagers at the school were using simplistic strategies to determine the ages of people they met online, often taking them at their word without exercising caution. He said this highlighted the urgent need for a "safety net of some sort".

It is hoped that an early working version of the software could be available by next summer.

Researchers believe the software could eventually be used to pick up on "stylistic footprints" used by paedophiles, which would help police track them as they move around the internet.

Police have been monitoring the impact of the pilot scheme, hoping the software will shorten the time-consuming process of examining logs of internet chat and other online material in the search for suspicious behaviour.

Alison Wilkinson, from Queen Elizabeth School, 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 software is being developed as part of a project funded by the Economic and Social Research Council and the Engineering and Physical Sciences Research Council.