Abstract

Humanity is in an age of technology where children have grown up with iPads, computers, and cell phones as part of their daily life. With the increased technology and communication, online grooming is a significant risk. Online grooming is when an adult instigates and escalates an emotional and often sexual relationship with a minor. Previous works have investigated methods to catch such predators, but often these solutions detect predators after the grooming process has begun. In this work, the sexual predator detection problem is examined to identify predators early in the grooming process. This work utilizes the preexisting early sexual predator detection (eSPD) framework by Vogt et al. [54], where the goal is to catch a predator as early and accurately as possible, along with the BigBird model to complete that goal. This work matches the current state-of-the-art results with an f-1 score of 0.89 or 89 percent.

Advisor

Montelione, Thomas

Department

Computer Science

Publication Date

2023

Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis

Download

Share

COinS
 

© Copyright 2023 Haley R. Bloom