Exploring high-level features for detecting cyberpedophilia

In this paper, we suggest a list of high-level features and study their applicability in detection of cyberpedophiles. We used a corpus of chats downloaded from www.perverted-justice.com and two negative datasets of different nature: cybersex logs available online and the NPS chat corpus. The SVM classification results show that the NPS data and the pedophiles’ conversations can be accurately discriminated with character n-grams, while in the more complicated case of cybersex logs high-level features significantly outperform the low-level ones and achieve a 97% accuracy.