Doctor of Philosophy (Ph.D.), Bowling Green State University, 2012, Communication Studies

Microlending through online venues has introduced a new model of lending through web 2.0 communication technologies. I examined micro lending through online venues – such as kiva.org, MicroPlace.com, and ACCION.org. The theoretical framework is based in Critical Cyberculture Studies and Critical Development Communication using visual analysis (Brummet, 2010; 2011; Mirzoeff, 2009; Nakamura, 2008; Olsen, 2007; Sosale, 2007) as my method, which is supplemented with interviews. I draw in part from visual rhetoric to inform my critique of the interplay of visual images, symbols, texts, and other elements in the microfinance web sites. On the home pages of Kiva.org, ACCION.org and MicroPlace.com, I analyzed the layout, including visuals and texts on their respective homepages. I examined the communication processes in these web 2.0 portals, because while some sites may indeed empower the poor, other sites may be disempowering to the poor. Kiva, ACCION, and MicroPlace thus reproduce issues of race, identity, and representation online, becoming discursive and rhetorical spaces where race and identity are produced and reproduced in various forms (Nakamura, 2002). Understanding the representations of third-world identities/bodies on micro lending sites is important. Also, global development initiatives such as kiva.org, MicroPlace.com, and ACCION.org have wide reaching ramifications; thus, the notion of empowerment of the poor, as reflected on the web portals of kiva.org, MicroPlace.com, and ACCION.org, bears scrutiny.

Committee: Radhika Gajjala PhD (Committee Chair); Lynda Dixon PhD (Committee Member); Ellen Gorsevski PhD (Committee Member); Shannon Orr PhD (Other) Subjects: African Studies; Banking; Black Studies; Business Education; Communication; Economics; Entrepreneurship; Ethnic Studies; Gender; Gender Studies; Health; Labor Economics; Mass Communications; Mass Media; Sub Saharan Africa Studies

Doctor of Philosophy (Ph.D.), Bowling Green State University, 2024, Media and Communication

This study utilizes an experimental design to explore how different frames impact individuals' perceptions of Muslim women when portrayed in news coverage of protests from the Middle East and North Africa region. Specifically, this research investigates the influence of news media frames on U.S. public perceptions of Muslim women activists, focusing on the impact of the hijab to test various perspectives related to minorities, gender, and stereotypical representations. In addition to examining the effect of text (positive and negative frames) and visuals (no visuals, visuals featuring veiled Muslim women, and visuals of unveiled Muslim women) on perceptions, this study also analyzes the influence of the interaction effect of the text and visual frames. This dissertation employed a factorial design, utilizing Amazon Mechanical Turk (MTurk) to conduct an online experiment. Participants were exposed to different news frames describing protests to assess their perceptions of Muslim women activists. One of the key findings of this study highlights the influence of Western-centric notions on perceptions of Muslims. Results demonstrated that positive text frames accompanied by visuals featuring unveiled women facilitated more positive implicit perceptions compared to negative frames. However, exposure to visuals featuring veiled women fostered more support toward Muslim women's protests compared to those exposed to unveiled visuals, regardless of whether the text frame is positive or negative. Additionally, results exhibited that preexisting stereotypes of oppression and victimization, along with interactions with Muslims, emerged as the most influential predictors in shaping perceptions. iv Based on the results, the author urges editors and journalists to carefully consider the goal of their coverage of protest news from the Middle East to ensure accurate and balanced portrayals that contribute to greater social inclusion, diversity, and equity in media discourse. The (open full item for complete abstract)

Committee: Louisa Ha PhD (Committee Chair); Kefa Otiso PhD (Other); Lara Langel PhD (Committee Member); Yanqin Lu PhD (Committee Member) Subjects: Mass Media; Middle Eastern Studies; Minority and Ethnic Groups; Womens Studies

Doctor of Philosophy, The Ohio State University, 2023, Electrical and Computer Engineering

The advent of deep neural networks (DNNs) have brought exciting new possibilities for the realization of automated driving functions. These data-driven methods have been widely applied to various driving tasks, including end-to-end urban driving. However, the use of these methods beyond simulated tests remains limited due to two significant shortcomings: (i) the lack of model transparency and (ii) the difficulty of generalizing beyond the training distribution. This dissertation aims to investigate methods for addressing the transparency and safety mitigation of learning-based controllers, specifically deep reinforcement learning (DRL) methods, to enable safe and predictable driving. To enhance interpretability, an interpretable and causal state representation, coined the driving forces, is proposed. This representation captures the causal relationship between the state and the produced control action by leveraging force features to encode the influence of internal and external factors on the ego vehicle. By training a DRL agent on this representation within a highway driving environment, the ability of the driving forces to encode and interpret the state-action causalities was demonstrated. Furthermore, an alternative paradigm for online adaptation by modifying the formulation of the driving forces is proposed to mitigate the behavior of the ego vehicle. The results showed that the ego vehicle was successful in mitigating its behavior and following desired new and unseen behaviors, without requiring modification to the underlying DNN. To address model transparency in black-box DNN-based driving policies, a knowledge distillation framework that combines interpretable decision trees with rule learning algorithms is proposed. This framework learns decision rule sets that represent the decision boundaries of the original driving policy. The driving forces are utilized to abstract the original state representation and ensure the interpretability of the learned explanati (open full item for complete abstract)

Committee: Umit Ozguner (Advisor); Keith Redmill (Committee Member); Qadeer Ahmed (Committee Member); Gladys Mitchell (Committee Member) Subjects: Artificial Intelligence; Computer Engineering; Electrical Engineering; Transportation