This article reports on an examination of the technology-facilitated abuse, cyberstalking, and the innovations gained by the authors’ research, including their development of an automated ability to identify and collect a complete set of federally prosecuted cyberstalking cases, and the use of natural language processing and regression methods to code and analyze those court records.
Among the new forms of technology-facilitated abuses, cyberstalking has become a growing and important problem. Cyberstalking involves the use of technology to stalk, threaten, or harass one or more individuals. For example, it can include tracking and intimidating a victim over social media, email, or text messages, or threatening to expose someone’s intimate photographs (sextortion). Cyberstalking has become a mechanism used by current or former domestic or intimate-partners, lone perpetrators, individuals targeting victims based on their employment or public image, and members of extremist groups. The innovations of this research are twofold. First, using multiple data sets, the authors developed an automated capability to identify and collect the complete set of all federally prosecuted cyberstalking cases in the U.S.. Second, the authors employ natural language processing, network, and regression methods to code and analyze the court records. The authors apply these methods in order to answer three main research questions: how many federal cyberstalking cases are there?; what kinds of stalking behavior are being committed?; and what characteristics are correlated with conviction and severity of punishment? (Published Abstract Provided)