ORCID Identifier(s)

0009-0007-8010-0544

Graduation Semester and Year

Summer 2024

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Computer Science

Department

Computer Science and Engineering

First Advisor

Chengkai Li

Second Advisor

Gautam Das

Third Advisor

Won Hwa Kim

Fourth Advisor

Shirin Nilizadeh

Abstract

The ubiquity of social media has transformed it into a rich source for reflecting people's opinions, behaviors, and interactions. Users frequently encounter factual claims in news, stories, and political statements, which can be either true or false. These claims significantly shape people's minds and behaviors, influencing not only individual perspectives but also broader public discourse. This study explores individuals' behaviors and perceptions toward factual claims by leveraging the concept of "check-worthiness" to analyze the relationship between such claims and user behaviors across datasets containing tens of millions of social media posts, particularly tweets from the platform X (formerly Twitter). It addresses key questions, including whether users exhibit different posting tendencies based on the check-worthiness of claims, the underlying reasons for these differences, and whether users are more likely to engage with content that aligns with the check-worthiness levels of their own posts.

Furthermore, the research introduces Wildfire, an innovative social sensing platform that empowers laypersons to conduct social sensing tasks using Twitter data without requiring programming or data analytics skills. Unlike existing tools that rely on simple keyword-based searches, Wildfire employs a heuristic graph exploration method to selectively expand the collected tweet-account graph, enhancing the collection of task-relevant data. This platform also offers a range of analytic tools, such as text classification, topic generation, and entity recognition, facilitating tasks such as trend analysis and public opinion sensing.

In addition to these methodological advancements, the research includes several real-world case studies that contribute to understanding the surveillance and impact of factual claims on specific topics, particularly in the contexts of the COVID-19 pandemic and climate change discussions on social media. Utilizing large language models, the study matches tweets with curated facts and misinformation, analyzing their stances and spatio-temporal spread. The findings highlight trends in misinformation during the COVID-19 pandemic and reveal the public's general tendency to believe claims regardless of their veracity.

In summary, the integration of social media into our daily lives has profoundly transformed how we interact with information, necessitating sophisticated approaches to understanding and managing the vast array of factual claims circulating online. The rise of platforms like Twitter has amplified the impact of these factual claims on public opinion and major societal events. By adopting the innovative approach of claim sensing, this study aims to bridge the gap between factual claims and human behavior.

Keywords

Claim sensing, Factual claim, Social media analytics

Disciplines

Computer Sciences | Data Science

Available for download on Tuesday, August 26, 2025

Share

COinS