Abstract

Computational risk detection has the potential to protect especially vulnerable populations from online victimization. Conducting a comprehensive literature review on computational approaches for online sexual risk detection led to the identification that the majority of this work has focused on identifying sexual predators after-the-fact. Also, many studies rely on public datasets and third-party annotators to establish ground truth and train their algorithms, which do not accurately represent young social media users and their perspectives to prevent victimization. To address these gaps, this dissertation integrated human-centered approaches to both creating representative datasets and developing sexual risk detection machine learning models to ensure the broader societal impacts of this important work. In order to understand what and how adolescents talk about their online sexual interactions to inform study designs, a thematic content analysis of posts by adolescents on an online peer support mental health was conducted. Then, a user study and web-based platform, Instagram Data Donation (IGDD), was designed to create an ecologically valid dataset. Youth could donate and annotate their Instagram data for online risks. After participating in the study, an interview study was conducted to understand how youth felt annotating data for online risks. Based on private conversations annotated by participants, sexual risk detection classifiers were created. The results indicated Convolutional Neural Network (CNN) and Random Forest models outperformed in identifying sexual risks at the conversation-level. Our experiments showed that classifiers trained on entire conversations performed better than message-level classifiers. We also trained classifiers to detect the severity risk level of a given message with CNN outperforming other models. We found that contextual (e.g., age, gender, and relationship type) and psycho-linguistic features contributed the most to accurately detecting sexual conversations. Our analysis provides insights into the important factors that enhance automated detection of sexual risks within youths' private conversations.

Notes

If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu

Graduation Date

2022

Semester

Spring

Advisor

Wisniewski, Pamela

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering and Computer Science

Department

Computer Science

Degree Program

Computer Science

Format

application/pdf

Identifier

CFE0009454; DP0027177

URL

https://purls.library.ucf.edu/go/DP0027177

Language

English

Release Date

November 2022

Length of Campus-only Access

None

Access Status

Doctoral Dissertation (Open Access)

Download

Share

COinS