Publication Date

5-2025

Date of Final Oral Examination (Defense)

2-24-2025

Type of Culminating Activity

Dissertation

Degree Title

Doctor of Philosophy in Computing

Department

Computer Science

Supervisory Committee Chair

Francesca Spezzano, Ph.D.

Supervisory Committee Member

Edoardo Serra, Ph.D.

Supervisory Committee Member

Nasir Eisty, Ph.D.

Abstract

Recommender systems play a crucial role in social media platforms by determining the order of information users see and aiding in news discovery. However, they can also cause harm through algorithmic privilege, excessive personalization, filter bubbles, and the recommendation and spread of misinformation. This dissertation aims to enhance proactive interventions against misinformation by addressing three key objectives. First, we investigate how recommendation algorithms contribute to misinformation propagation in social networks and propose a framework to measure their impact, introducing a novel information diffusion model that incorporates network, news, and user features. Second, we examine misinformation intervention strategies, from fake news detection and removal to social science-based approaches, assessing their robustness against circumvention and addressing the amplification of misinformation caused by recommender algorithms. Third, we explore towards the development of a misinformation-aware recommender system that prioritizes factors such as trustworthy networks and trustworthy neighbors. Furthermore, we analyze the interplay between news diversity in recommendations and the spread of misinformation. The outcomes of this research strengthen misinformation intervention strategies and provide critical insights for social scientists in fostering a more resilient and informed digital ecosystem.

Comments

ORCID, 0000-0002-5030-4841

DOI

https://doi.org/10.18122/td.2351.boisestate

Download

Available for download on Saturday, May 01, 2027

Share

COinS