Abstract

The internet plays an ever-increasing part in the day-to-day lives of many people. Ubiquitous computing has given rise to sophisticated, streamlined and faster connections across a range of devices. Mobile smart phones are in the hands of children as young as five years old, and whilst this allows them to interact with educational applications and the wealth of information available on-line, it can put them in danger.
There has been a consistent stream of stories involving children and adolescents being at risk because of unsafe on-line behaviour. Predators can prey on the vulnerable, by pretending to be a peer and convincing them, by charm or threats, to compromise their safety. Governments across the globe have initiatives to combat this threat, there are working groups and police task forces in place to respond to both the growing number, and impact of these incidents on children, young people, families and communities. In order to monitor on-line conversation and identify different levels of threats, the SafeChat system was designed and implemented using an ontology-based system and Natural Language Processing (NLP) techniques.