Authorship Attribution of Predators in Chat Conversations
Primary Investigator:
Julia (Taylor) Rayz
Kanishka Misra, Hemanth Devarapalli, Julia Taylor Rayz
Abstract
Authorship Attribution (AA) of written content presents several advantages within the digital forensics domain. While AA has been successful when applied to long documents, recent works have shown improved performance of neural AA models on short texts such as tweets and online conversations (Schwartz et al., 2013; Ruder et al., 2016; Shrestha et al., 2017). Concurrently, the rise of social media as well as a plethora of chat messaging platforms have made it easier for teenagers to be vulnerable to online predators. In this work, we present a new model to attribute authors to messages from a corpus consisting of chat conversations, some of which involve online predators, and perform subsequent analysis of neural representations of messages. Our results show comparable performance to prior work for Authorship Attribution and highlight differences between predatory and non-predatory styles.