Can We Automate the Analysis of Online Child Sexual Exploitation Discourse?

TL;DR

This study explores the potential of automated natural language processing methods to identify predatory behaviors in online chats between minors and adults, aiming to replace manual expert annotation and improve online child safety monitoring.

Contribution

It introduces a framework for automatically detecting grooming behaviors in chat messages, informed by psychological theories, and evaluates the performance of NLP models against human annotations.

Findings

Models show consistent classification of behaviors

Automated methods outperform random baselines

Approach aligns with human annotations but needs improvement

Abstract

Social media's growing popularity raises concerns around children's online safety. Interactions between minors and adults with predatory intentions is a particularly grave concern. Research into online sexual grooming has often relied on domain experts to manually annotate conversations, limiting both scale and scope. In this work, we test how well-automated methods can detect conversational behaviors and replace an expert human annotator. Informed by psychological theories of online grooming, we label $6772$ chat messages sent by child-sex offenders with one of eleven predatory behaviors. We train bag-of-words and natural language inference models to classify each behavior, and show that the best performing models classify behaviors in a manner that is consistent, but not on-par, with human annotation.