Theses and Dissertations from UMD
Permanent URI for this communityhttp://hdl.handle.net/1903/2
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
31 results
Search Results
Item NATURAL LANGUAGE PROCESSING, SOCIAL MEDIA, AND EPIDEMIC MODEL-ING FOR WILDFIRE RESPONSE AND RE-SILIENCE ENHANCEMENT(2024) Ma, Zihui; Baecher, Gregory B; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Effective disaster response is critical for communities to remain resilient and advance the development of smart cities. Responders and decision-makers benefit from reliable, timely measures of the issues impacting their communities during a disaster, and social media offers a potentially rich data source. Social media can reflect public concerns and behaviors during a disaster, offering valuable insights for decision-makers to understand evolving situations and optimize resource allocation. A comprehensive literature review of natural language processing (NLP) of social media data in disaster management, covering 324 articles published between 2011 and 2022, revealed a gap in applying NLP techniques to wildfire scenarios. Meanwhile, the increasing frequency of wildfires highlights the need for advanced management tools. To address this, we integrated the BERTopic and SIR models to capture public responses on Twitter during the 2020 western U.S. wildfire season, analyzing both the magnitude and velocity of topic diffusion. The results displayed a clear relationship between topic trends and wildfire propagation patterns. The parameters estimated from the SIR model for selected cities revealed that residents expressed various levels of concern or demand during wildfires. The study also demonstrated a practical framework for utilizing social media data to aid wildfire evacuations. Through social network analysis, we clarified the roles of key information disseminators and provided guidelines for extracting high-priority information. Although biases in social media and model limitations exist, the study offers qualitative and quantitative approaches to investigate wildfire response and sup-port community resilience enhancement.Item NAME, IMAGE AND LIKENESS: A CONTENT ANALYSIS OF HOW WOMEN STUDENT ATHLETES SHARE THEIR STORIES AND LIVED EXPERIENCES ON SOCIAL MEDIA IN THE AGE OF NIL(2023) Scovel, Shannon Marie; Oates, Sarah A; Journalism; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This dissertation assesses the self-representation and representation of ten elite collegiate women athletes during the first year of the NCAA’s new ‘name, image and likeness’ policies. Building on theories of representation, gender performance, self-presentation and intersectionality, this study explores how women athletes reproduce notions of feminism, femininity and athleticism on their public TikTok, Instagram and Twitter accounts. Each of the women in this study have at least 50,000 followers across their social media accounts, and the content they produced on these platforms over the 12-month period from July 1, 2021, to July 1, 2022, serves to both reflect and reject hegemonic norms surrounding women in sport. Previous research has demonstrated that women athletes remain marginalized and underrepresented in sports. Scholars have also noted that women athletes typically represent themselves on social media in ways that highlight their personal lives, as opposed to their athletic experiences. This study explores these questions of self-representation through a content analysis of social media posts produced by ten collegiate women and addresses how these women navigated digital storytelling within the neoliberal, capitalist, patriarchal U.S. college sports media ecosystem. The ways in which athlete content was reproduced by journalists during this same period was also assessed. Findings show that journalists rarely engaged with women athletes’ posts during the first year of the NCAA’s new NIL policies and presented women’s success in the NIL era as surprising, unexpected and unrelated to athletic achievements. This dissertation adds to the larger body of research on women’s representation and self-representation in sports but adds a new dimension to this subject by exploring such representations in the collegiate environment, an arena in which athletes were previously denied the opportunity to earn money from their digital storytelling and online brands. The ways in which women challenge and reproduce hegemonic norms in their social media content during this period also contributes to the broader understanding of gender tensions in sports.Item Psychological inoculation against vaccine misinformation: why and how it works(2023) Wang, Yuan; Nan, Xiaoli; Communication; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Vaccine misinformation has posed a significant threat to public health. Drawing upon inoculation theory, this dissertation investigates whether exposure to an inoculation message – a message that forewarns and refutes potential persuasive attacks – can confer resistance to misinformation about COVID-19 vaccines. Based on two online experiments, this research seeks to answer four overarching questions: Can exposure to an inoculation message reduce susceptibility to misinformation? Through which mechanisms does inoculation message confer resistance to misinformation? Does the effect of inoculation messages vary among initially informed, uninformed, and misinformed individuals? How do partisan source cues (in-group vs. out-group) impact the effectiveness of inoculation messages among politically affiliated individuals? Study 1 investigated the effectiveness, mechanisms, and recipient factors related to inoculation messages. A two-condition (inoculation vs. control) between-subject experiment was conducted (N = 659). Results indicated that exposure to an inoculation message effectively reduced individuals' susceptibility to misinformation. Inoculation message not only counteracted beliefs in misinformation but also protected positive attitudes and intentions toward COVID-19 vaccination. Moreover, perceived ease of counterarguing and anger were identified as significant mediators underlying the persuasive effects of the inoculation message, while counterarguing was not a significant mediator. Furthermore, the effectiveness of inoculation message remained consistent among initially informed, uninformed, or misinformed groups, suggesting that inoculation message offers both prophylactic and therapeutic effects. Study 2 examined how partisan source cues impacted inoculation message effectiveness. A 2 (in-group vs. out-group inoculation) X 2 (in-group vs. out-group misinformation) between-subject online experiment was conducted among politically affiliated individuals (N = 448). Results showed no main or interaction effects of in-group (vs. out-group) inoculation and in-group (vs. out-group) misinformation on persuasive outcomes, suggesting that the efficacy of inoculation messages in conferring resistance to misinformation did not differ based on whether the inoculation or misinformation messages came from an in-group or out-group source. Additionally, party identification strength moderated the impact of in-group (vs. out-group) inoculation on beliefs in COVID-19 vaccine misinformation and COVID-19 vaccination attitudes. Surprisingly, the advantage of in-group inoculation over out-group inoculation was stronger among individuals with lower levels of party identification. Moreover, out-group inoculation appeared to be more persuasive than in-group inoculation among individuals with extremely strong political identification. This dissertation offers several theoretical and practical implications for health communication research and practice. First, this research contributes to inoculation theory by examining two alternative mechanisms – perceived ease of counterarguing and anger – underlying inoculation message effects. The findings underscore the importance of considering cognitive, meta-cognitive, and affective routes that underlie resistance to persuasion. Additionally, this research expands the scope of inoculation theory by demonstrating its effectiveness among initially informed, uninformed, and misinformed individuals. These results suggest that inoculation messages can be useful beyond the traditional scope of cultural truisms, offering both prophylactic and therapeutic effects. Furthermore, the study challenges the conventional assumption that messages from in-group sources are more persuasive than those from out-group sources, indicating that political groups should work together to address vaccine hesitancy. Overall, this dissertation supports the use of inoculation messages as an effective tool in counteracting misinformation and promoting vaccination acceptance.Item Towards Multimodal and Context-Aware Emotion Perception(2023) Mittal, Trisha; Manocha, Dinesh Dr.; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Human emotion perception is a part of affective computing, a branch of computing that studies and develops systems and devices that can recognize, interpret, process, and simulate human affects. Research in human emotion perception, however, has been mostly restricted to psychology-based literature which explores the theoretical aspects of emotion perception, but does not touch upon its practical applications. For instance, human emotion perception plays a pivotal role in an extensive array of sophisticated intelligent systems, encompassing domains such as behavior prediction, social robotics, medicine, surveillance, and entertainment. In order to deploy emotion perception in these applications, extensive research in psychology has demonstrated that humans not only perceive emotions and behavior through diverse human modalities but also glean insights from situational and contextual cues. This dissertation not only enhances the capabilities of existing human emotion perception systems but also forges novel connections between emotion perception and multimedia analysis, social media analysis, and multimedia forensics. Specifically, this work introduces two innovative algorithms that revolutionize the construction of human emotion perception models. These algorithms are then applied to detect falsified multimedia, understand human behavior and psychology on social media networks, and extract the intricate array of emotions evoked by movies. In the first part of this dissertation, we delve into two unique approaches to advance emotion perception models. The first approach capitalizes on the power of multiple modalities to perceive human emotion. The second approach leverages the contextual information, such as the background scene, diverse modalities of the human subject, and intricate socio-dynamic inter-agent interactions. These elements converge to predict perceived emotions with better accuracy, culminating in the development of context-aware human emotion perception models. In the second part of this thesis, we forge connections between emotion perception and three prominent domains of artificial intelligence applications. These domains include video manipulations and deepfake detection, multimedia content analysis, and user behavior analysis on social media platforms. Drawing inspiration from emotion perception, we conceptualize enriched solutions that push the conventional boundaries and redefine the possibilities within these domains. All experiments in this dissertation have been conducted on all state-of-the-art emotion perception datasets, including IEMOCAP, CMU-MOSEI, EMOTIC, SENDv1, MovieGraphs, LIRIS-ACCEDE, DF-TIMIT, DFDC, Intentonomy, MDID, and MET-Meme. In fact, we propose three additional datasets to this list, namely GroupWalk, VideoSham and IntentGram. In addition to providing quantitative results to validate our claims, we conduct user evaluations where applicable, serving as a compelling testament to the remarkable outcomes of our experiments.Item Modeling Wise Angers Online: Generation Z Activists and Their Digital Rhetorics of Feminist Rage(2023) Starr, Brittany Noelle Schoedel; Enoch, Jessica; English Language and Literature; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)“Modeling Wise Angers Online: Generation Z Activists and Their Digital Rhetorics of Feminist Rage” works at the nexus of feminist theory, digital media studies, and rhetoric to investigate how teen and young adult activists use 21st century social media technologies to challenge the sexist, racist, ageist, and ableist anger norms that disenfranchise young women in the public sphere. Each chapter theorizes what I call a “wise anger” strategy that its principal subject deploys to generate rhetorical agency for angry girl activists and change oppressive anger norms. The activists I examine are Greta Thunberg, Thandiwe Abdullah, and Shina Novalinga. While their causes range from the climate crisis to racial justice and Indigenous rights, and their primary platforms in my case studies are Twitter, Instagram, and TikTok, respectively, they all make innovative, strategic use of digital affordances to reframe young women’s anger in public discourse. Examining datasets I compiled from the activists’ social media posts between 2018-2022, I use grounded theory and rhetorical analysis to identify patterns in the anger expressions in the multimodal, multilayered posts. I read the patterns through feminist and Black feminist theories of oppressive anger norms (Jaggar, Ahmed, Traister, Chemaly, Lorde, Cooper, Judd, Collins), cultural rhetorical frameworks (Powell et al.; Karetak, Tester, and Tagalik) and youth activist rhetorical frameworks (Applegarth, Hesford, Taft, Dingo). This dissertation is premised on the understanding that emotions have a biological basis, but are constructed socially, rhetorically, and culturally and thus tend to be scripted in ways that reproduce asymmetrical relations of power (Aristotle, Dixon, Fine, Gross, Harrington, Koerber). Ultimately, I develop a theory of wise anger as an angry response to injustice that is intelligent, informed, constructive, justice-oriented, hope-driven, rational, reasonable, and moral. The wise anger these youth activists model through their digital rhetorics on social media is part of a genealogy of feminist rage that envisions and enacts a more inclusive, more livable world.Item USING SOCIAL MEDIA AS A DATA SOURCE IN PUBLIC HEALTH RESEARCH(2022) Sigalo, Nekabari; Frias-Martinez, Vanessa; Library & Information Services; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Researchers have increasingly looked to social media data as a means of measuring population health and well-being in a less intrusive and more scalable manner compared to traditional public health data sources. In this dissertation, I outline three studies that leverage social media as a data source, to answer research questions related to public health and compare traditional public health data sources to social media data sources. In Study #1, I conduct a study with the aim of developing, from geotagged Twitter data, a predictive model for the identification of food deserts in the United States, using the linguistic constructs found in food-related tweets. The results from this study suggest the food-ingestion language found in tweets, such as census-tract level measures of food sentiment and healthiness, are associated with census tract-level food desert status. Additionally, the results suggest that including food ingestion language derived from tweets in classification models that predict food desert status improves model performance when compared to baseline models that only include socio-economic characteristics. In Study #2, I evaluate whether attitudes towards COVID-19 vaccines collected from the Household Pulse Survey can be predicted using attitudes extracted from Twitter. The results reveal that attitudes toward COVID-19 vaccines found in tweets explain 61-72% of the variability in the percentage of HPS respondents that were vaccine hesitant or compliant. The results also reveal significant statistical relationships between perceptions expressed on Twitter and in the survey. In Study #3, I conduct a study to examine whether supplementing COVID-19 vaccine uptake forecast models with the attitudes found in tweets improves over baseline models that only use historical vaccination data. The results of this study reveal that supplementing baseline forecast models with both historical vaccination data and COVID-19 vaccine attitudes found in tweets reduce RMSE by as much as 9%. The studies outlined in this dissertation suggest there is a valuable signal for public health research in Twitter data.Item Féminisme sur Instagram : un bilan mitigé(2022) Danos, Clara Clémentine Alice; Orlando, Valérie K; French Language and Literature; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Online feminism was birthed alongside the Internet in its infant stages. With the years, it evolved into a virtual fight on social media and especially on Instagram. The goal of this thesis is to analyze content created by French feminists on Instagram and decide if it could be identified as a fourth wave of feminism in which women would rule the virtual world emancipating themselves from patriarchy in virtual life, in hopes of a more equitable society offline. Presently, alterations in combat against this ubiquitous foe are becoming more accessible, pedagogical, and aesthetic. However, these adaptations corrupt the core of feminism itself; lost consistence in the process with a lack of references, novelty, and anti-capitalist spirit. These inhibitors actively preventing the progression of a fourth wave. Consequently, feminists currently navigate the parameters of male engineered social media and experiencing an Instagram that is complicit in masculinist abuse through internal politics and outside actors.Item Creepy or Cool? An Exploration of Non-Malicious Deepfakes Through Analysis of Two Case Studies(2022) Cleveland, Keaunna; Shilton, Katie; Library & Information Services; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Several studies have examined the harms associated with the development of deepfake technology and its use by malicious actors, but less research has been devoted to deepfakes created by non-malicious creators and the ways people react to deepfakes developed without malicious intent. This study attempts to close this research gap through the exploration of two case studies that demonstrate non-malicious deepfake use on Instagram and Twitter. Using sensemaking, privacy as contextual integrity, and audience theory to guide the analysis of publicly available posts, tweets, and records, this study examines how people interact with and react to non-malicious deepfakes online. Building on these findings, this thesis suggests how social media platforms might integrate signifiers in their design that afford sensemaking for those interacting with deepfake technology and discusses how ethical frameworks and practices from values-oriented design and value-based engineering in design may help guide creators as they develop deepfake technology videos and applications for non-malicious purposes.Item TALKING ABOUT JUSTICE: PREDICTING ACTOR ENGAGEMENT ON SOCIAL MEDIA AFTER A GALVANIZING EVENT(2021) Glasgow, Kimberly Ann; Vitak, Jessica; Library & Information Services; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Social media contributes to discourse around and framing of major societal issues, and enables community formation, social change, and activism. It provides opportunities to engage in discourse, gain and share knowledge, and form ties with others around an issue, topic, or cause. This dissertation explores how justice, an important concept underlying social systems, is expressed in Twitter data in the context of high-salience, galvanizing local events, and leverages that information to predict whether newcomers to the issue will continue their digital engagement on the topic over time. It also attempts to quantify whether, and how much, a set of factors or dimensions previously associated with engagement in the physical realm contribute to digital engagement. These dimensions—identity, emotion, effort, and social embeddedness—are informed by prior work on social movements, digital activism, and related fields. Rather than rely on hashtags, this dissertation uses machine learning to detect justice-related Twitter activity. This advance in methods provides a richer understanding of discourse around a complex, multifaceted topic like justice. It allows deeper insight into the social media activity of newcomers to the justice community, and the networks they are embedded in. The approach is developed and applied first to Twitter data from Baltimore around the 2015 death of Freddie Gray from injuries sustained while in police custody, and the protests and riots that followed in Baltimore. To test for generalizability, the same approach is then applied to a second dataset, collected from Cleveland at the time of the death of Tamir Rice, who was shot and killed by police in 2014. Findings show that digital engagement in justice discourse on social media can be predicted, based on aspects of social embeddedness, emotion, and effort. To the degree that committed individuals are at the heart of social movements and efforts to spur social and civic change, and forming and being embedded in appropriate network structures is critical for channeling commitment into action and eventual success, this work contributes to greater understanding of these phenomena. Findings from this research could contribute to the design of technology to support civic engagement through social media platforms.Item From Oversharing to Sharenting: How Experts Govern Parents and Their Social Media Use(2021) Kumar, Priya; Vitak, Jessica; Library & Information Services; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)A newborn swaddled in a parent’s arms. A kindergartner posing on the first day of school. Such images, commonly found in family photo collections, now regularly appear on social media. At the same time, public discourse asks if—or sometimes asserts that—posting images online might put children’s privacy, dignity, and autonomy at risk. Prior research has documented the pressure, scrutiny, and judgment that parents, especially mothers, endure. It seems that parents’ use of social media is yet another cause for concern. How did this happen? This dissertation examines how power, manifesting as expertise, works through three fields of discourse to govern parents’ social media conduct. Grounding this project in post-structuralist epistemology, I study this question using the analytical technique of governmentality, which is a means of tracing how authorities intervene in the lives of individuals. First, I illustrate how a specific site of social media expertise, the once-popular blog STFU, Parents, constructs the problem of “oversharing” as a form of inappropriate social media use. Second, I explain, how news media expertise constructs the problem of “sharenting,” a portmanteau of the words “share” and “parenting,” as a form of risk to children. Third, I discern how academic expertise obliges parents to govern their own social media conduct by appealing to their subjectivity. In each field of discourse, I observe how expertise frames parents’ social media conduct as a matter of individual responsibility, even though much of what happens to information online lies outside individual control. I use this analysis to suggest future directions for research on social media and privacy that goes beyond the gendered public/private boundary and engages with the world as a site of entangled relations rather than individual entities.