Digital Commons Network^™

Facilitating Deep Learning Performance Analysis Through Automated Roofline Model Generation, Irvin Lopez-Audetat

Open Access Theses & Dissertations

This thesis presents a tool to profile deep learning (DL) and machine learning (ML) models by collecting FLOPs, memory movement, and timing data through cyPAPI to generate roofline performance models. The tool is containerized for portability and reproducibility, integrates directly with PyTorch workflows, and provides fine grained insights into computational bottlenecks across model components. Unlike prior system-level or benchmarking-centric tools, this project empowers developers and researchers with an accessible, modular framework for performance analysis and optimization.

Efficient Adaptive Spline-Based Path Planning For In-Space Servicing, Assembly, And Manufacturing Applications, Christian Lozoya

Open Access Theses & Dissertations

Autonomous robotic systems operating in cluttered and partially observed environments require trajectory generation methods that produce smooth and dynamically feasible motion while reacting to locally sensed obstacles. This requirement is especially pronounced for free-flyer and in-space servicing, assembly, and manufacturing (ISAM) platforms, where onboard sensing is sparse, global environmental information is unavailable, and communication or computational resources are constrained. In such settings, motion plans must be updated online using incomplete and rapidly changing local observations, while avoiding excessive replanning that can lead to oscillatory or unstable behavior. Many existing approaches either rely on dense optimization over extended horizons, which is …

Further Insights Into The Network Link Outlier Factor's (Nlof) Light-Load Penalty, Sunday Oluwaleke Ogundele

Open Access Theses & Dissertations

This research investigates the performance of the Network Link Outlier Factor with Most Likely Links (NLOF:MLL), under varying network load conditions. Earlier studies reported that the NLOF:MLL algorithm experienced a noticeable drop in fault-localization accuracy when operating in lightly loaded networks. To further examine this limitation, 240 experiments were carried out to observe how the algorithm responds as overall network load increases. The evaluation focused on the classification performance metrics: precision, recall, and F1-score. The results show that NLOF:MLL’s effectiveness improves as network load increases but that the rate of improvement slows progressively, eventually stabilizing in a pattern consistent with …

Applying The Quadruple Process Model To A Measure Of Implicit Bias Towards Individuals With Alcohol Use Disorder, Erin Marie Portillo

Open Access Theses & Dissertations

Alcohol use disorders is one of the major burdens of disease, yet it remains one of the disorders with the lowest treatment prevalence. For individuals who seek treatment for their alcohol use disorder, they may experience higher levels of social disintegration, guilt, shame, and stigmatization. Stigma is a multistep process that makes way for stereotypes, prejudice, and acts of discrimination to take place. With labels such as “alcoholic” implying negative connotation and being frequently used, the fear of being stigmatized and negatively labeled often results in delayed or entirely avoided care. Research has turned to the use of implicit measures …

Lorentz Invariance And Quantum Coordination: A Neo-Aristotelian Framework For Entanglement And Relativity, Angel Rafael Sosa Muniz

Open Access Theses & Dissertations

The phenomenon of quantum entanglement has been at the center of heated debates among physicists and philosophers since the dawn of the quantum era. The early discussions initiated by Einstein, Podolsky, and Rosen—regarding potential violations of the Principle of Relativity by entangled particles, expressed in the so-called “EPR Paradox”—ultimately culminated in the demonstration of Bell’s theorem and its violations. Since then, philosophers of science and physicists have developed multiple proposals attempting to account for the ontological status of quantum formalism and its possible tension with relativistic principles. This thesis contributes to these efforts by advancing a Neo-Aristotelian ontological framework inspired …

Gpu-Based Electromagnetic Microwave Tomography For Brain Imaging And Stroke Detection, Pablo Sotelo Torres

Open Access Theses & Dissertations

Each year, an estimated 795,000 people in the U.S. suffer a stroke, with approximately 610,000 being first-time cases. Of these, 87% are ischemic strokes, while the remaining 13% are hemorrhagic. Current imaging methods, such as Computed Tomography (CT), Positron Emission Tomography (PET), and Magnetic Resonance Imaging (MRI), provide useful information into brain tissue properties; while each technique has its advantages, they remain expensive, non-portable, and often too slow for emergency bedside or in-ambulance use. Electromagnetic Microwave Tomography (EMT) offers a promising alternative: an affordable, portable, rapid, and safe method for stroke detection. By contrasting dielectric properties between healthy and affected …

Hubert-Based Models And Evaluation Strategies For Pragmatically-Faithful Speech To Speech Translation, Javier Vazquez

Open Access Theses & Dissertations

Pragmatic fidelity in speech-to-speech translation (S2ST) has largely been understudied, leading to communication tools inadequate to support non-superficial dialog. We aim to improve pragmatic faithfulness in English-Spanish translation through the development of machine learning models that are able to predict a corresponding pragmatic representation in the other language. To evaluate performance, we developed a pipeline that utilizes a recently-developed pragmatic similarity evaluation metric to compare models. Further, we developed models that exploit HuBERT features as these have been found suitable for various prosody and pragmatics related tasks. Our models outperformed human and state-of-the-art predictions, albeit the methodology being limited to …

A Pilot Study Investigating The Impact Of Cigarette Smoking On Oral Nitrate-Reducing Bacteria Within The Oral Microbiome, Tomas L Nicholas, Raul Bescos, Amazon Doble, Katie Muddiman, Paul Kellett, Abish S Stephen, Lisa Du Toit, Robert Witton, Zoë Ls Brookes

Peninsula Dental School

ObjectiveTo evaluate the impact of cigarette smoke on the activity and abundance of oral nitrate reducing bacteria (ONRB) in vivo and in vitro.Materials and MethodsAn observational study comprising of eleven cigarette smokers and thirty-one non-smokers. A saliva and nitrate-rinse sample were collected from each participant to analyse salivary biomarkers and the activity and abundance of ONRB in vivo. Additionally, the effect of cigarette smoke on the nitrate-reducing activity of four Rothia species was investigated in vitro.ResultsThe beta diversity of the oral microbiome was significantly different between smokers and non-smokers (P = 0.0063). Neisseria flavescens and Actinomyces oris were significantly higher …

Life Cycle Assessment Comparing Electric And Petrol-Powered Rigid Inflatable Boats For Recreational And Harbour-Master Applications, Amir Bordbar, Badr Moutik, Adam Kyte, Simon Ussher, Keri Collins, Marcin Roszkowski, Tony Wilson, Sarah Fear

School of Geography, Earth and Environmental Sciences

This study presents the first cradle-to-grave life cycle assessment (LCA), focusing exclusively on environmental impacts, comparing electric and petrol-powered rigid inflatable boats (RIBs) under low-usage recreational and high-usage harbour-master scenarios. The study follows ISO-compliant LCA methodology and emphasizes transparency in data and assumptions. Ten environmental impact categories were analysed, including global warming potential (GWP), particulate matter formation, acidification, eutrophication, ecotoxicity, and resource scarcity. Results show that switching to electric RIBs reduces GWP by 34 % for recreational and 52 % for harbour-master scenarios, despite higher production-phase impacts across all categories. The electric RIB delivers notable operational benefits in GWP and …

Reservoir Characterization Of The Clarksville Formation In Red River County, Texas, Matthew Allen Parker

Electronic Theses and Dissertations

The Upper Jurassic Clarksville Formation in northern Texas and southern Oklahoma is a productive but poorly understood sandstone–conglomerate unit with unresolved questions regarding its depositional environment and sediment provenance. This study evaluates the depositional setting, spatial distribution, and diagenetic controls of the Clarksville Formation using integrated core and well log analyses in Red River County, Texas. Three cores (Lennox Unit #1, Reeder Oil Unit #2, and Reeder Oil Unit #7) were analyzed for lithology, mineralogy, porosity, permeability, and fluid saturation, and correlated with 137 well logs to map formation thickness and regional trends.

Four lithofacies were identified based on core …

Studies Of Trispyrazolylborate Molybdenum-Oxo Compounds And Related Reactions, Miranda Sanders

Graduate Theses and Dissertations

This dissertation details the design, development, and synthesis of a molybdenum-dioxo complex bearing a trispyrazolylborate ligand. The system was tested for its efficiency as a catalyst for deoxydehydration with a small substrate scope. Deoxydehydration reactions with 1-phenyl-1,2-ethanediol, diethytartrate, and 1,2-octanediol show low yields with catalytic turnover only found for diethyltartrate. Electrochemical studies show well behaved PCET behavior with two distinct electron transfer events in the presence of a proton donor. Studies of two molybdenum-dioxo complexes bearing dianionic pincer ligands were conducted to expand further on systems that have been extensively studied in our lab. Deoxydehydration of glycerol by a molybdenum-dioxo …

Synthetic Dataset For Understanding Negation In Text-Guided Image Editing, Nhat-Tan Bui

Graduate Theses and Dissertations

Negation is a fundamental linguistic concept used by humans to convey information that they do not desire. Despite this, minimal research has focused on negation within text-guided image editing. This lack of research means that vision-language models (VLMs) for image editing may struggle to understand negation, implying that they struggle to provide accurate results. One barrier to achieving human-level intelligence is the lack of a standard collection by which research into negation can be evaluated. This thesis presents the first large-scale dataset, Negative Instruction (NeIn), for studying negation within instruction-based image editing. Our dataset comprises 366,957 quintuplets, i.e., source image, …

Modern Water And Sediment Transport Patterns In Alluvial Ridge Basins Of The Rio Grande Delta, South Texas, Usa, Waqid Nabi

Theses and Dissertations

Alluvial ridge (AR) basins are shallow, marine-influenced depressions within river deltas, yet their flow and sediment dynamics remain poorly understood. This study examines hydrodynamics and sediment transport patterns in the Bahia Grande Complex, which are AR basins of the Rio Grande Delta, using tilt current meters, water-level and turbidity loggers, and suspended and bed-sediment sampling. The AR basins function as wind-dominated, microtidal environments, where pre-existing basin geometry and hydrodynamic forcings control water flow and sediment distribution patterns. Seasonal wind patterns and storms modulate water levels and flow connectivity. Substrate patterns reflect frequent resuspension in the shallow northern basin, maintaining coarse …

Hiring Expectations For New And Future Workers In Animal Genetics Careers, Meghan C. Wulster-Radcliffe, James N. Macleod, Robert J. Coleman, John S. Radcliffe, Rebekah B. Epps, Carol D. Hanley, Cheyenne S. Summers, Steven A. Zinn, Sarah A. Reed

Animal and Food Sciences Faculty Publications

To better align educational outcomes with workforce demands in animal and equine genetics, this study surveyed 65 industry professionals on entry-level competencies, hiring trends, and graduate preparedness of undergraduate students. The survey assessed employer ratings on the importance and perceived proficiency of seven core competencies: critical thinking, communication, teamwork, leadership, professionalism, career/self-development, and equity/inclusion, as well as technical knowledge and broader skills relevant to animal breeding and genetics. Results revealed oral communication, teamwork, and work ethic were rated as the most critical transferable skills, though notable proficiency gaps persist, particularly in leadership and professionalism. Foundational technical knowledge in genetics, molecular …

Examining Grade 10 Students’ Ict Utilization In Bhutan, Gembo Tshering, Thinley Phuntsho

Journal of Global Education and Research

In today’s educational landscape, Information and Communication Technology (ICT) presents immense potential for reshaping pedagogical practices and equipping students with essential skills for the digital era. This study delves into assessing Grade 10 students’ ICT proficiency in Bhutan, a nation renowned for its dedication to Gross National Happiness (GNH) and sustainable development. The study also explores affective factors, including attitudes and confidence. Despite strides in promoting digital literacy and widening access to ICT resources, challenges persist in ensuring equitable access and meaningful utilization of technology, particularly at the secondary education level. Through a thorough analysis of survey data collected from …

Internationalization Of Indian Higher Education: Mapping The Origins And Academic Choices Of International Students, Sanjeev Kumar Jha

Journal of Global Education and Research

This study explores the geographical origins, academic preferences, and key factors influencing the enrollment of international students in Indian higher education institutions. Drawing on primary data from interviews with 25 international students and secondary data from All India Survey on Higher Education reports, it provides an in-depth analysis of student mobility trends and factors shaping their choices. The results indicate that the international student enrollment is significantly below the Indian government’s target of 500,000 by 2024. Students come from Nepal, Sri Lanka, Afghanistan, Bhutan, and Nigeria, with Karnataka and Punjab emerging as the most preferred states within India. Engineering, particularly …

Uncertainty Estimation For Graph-Based Learning In Digital Pathology, Saba Heidari Gheshlaghi, Nasim Yahyasoltani, Masoud Ganji

Computer Science Faculty Research and Publications

High-resolution digital scans of pathology slides, known as whole slide images (WSIs), have detailed spatial and contextual information for diagnosing cancer. However, the classification performance of WSIs by deep learning models is typically compromised by data with a different distribution, known as out-of-distribution (OOD), resulting in unreliable predictions. Therefore, having a reliable predictive uncertainty estimation is crucial for clinical adoption. This article comprehensively studies graph-based uncertainty estimation for WSI classification using two cutting-edge graph neural network (GNN) architectures: 1) graph attention networks (GAT); and 2) GraphSAGE. In this work, we introduce the first unified multihead GNN framework that leverages GraphSAGE …

The Transition Zone Hypothesis On Biodiversity Of The Indian River Lagoon System, Florida, Richard L. Turner

Ocean Engineering and Marine Sciences Faculty Publications

The Indian River Lagoon System (IRL) is regarded as having a high level of biodiversity. The level of biodiversity is often attributed in part to the 2-degree latitudinal span (251 km) of the IRL and to its position in a transitional zone between tropical and temperate biotic provinces. This hypothesis was tested for submerged aquatic vegetation (SAV) based on a recent treatise (Littler, Littler, and Hanisak, 2008). Five of seven species of seagrass are tropical, and two have either a broad or very narrow latitudinal distribution; the transition-zone hypothesis does not hold for seagrasses. Most macroalgae (131 spp.) are tropical/subtropical; …

Evaluation Of Ramp Instead Of Forage During Grain Adaptation And Different Sorghum Varieties On Enteric Methane Production From Cattle, Jack T. Fort

Department of Animal Science: Dissertations, Theses, and Student Research

Two experiments were conducted to evaluate the use of RAMP (Cargill Corn Milling, Blair, Nebraska) compared with traditional forage-based adaptations on methane (CH₄) and carbon dioxide (CO₂) emissions, performance, and carcass characteristics. Experiment 1, 64 yearling British × Continental steers (initial BW = 423 ± 6 kg) were adapted using 100% RAMP (RAMP) or a traditional forage diet (CON; 43% forage) before transitioning 22 d to a common finishing diet containing 65.6% steam-flaked corn. Emissions were measured in paired replications using a two-chamber indirect calorimetry barn. During adaptation RAMP did not affect CH₄ production (g/d; …

Integrating Innovative Sensing And Ground Measurements For Forage Mass Estimation In Rye Pasture, Pedro Henrique Jota Fernandes

Department of Animal Science: Dissertations, Theses, and Student Research

The United States beef industry relies on forages to provide up to 80% of total lifetime dry matter intake, making accurate forage mass estimation essential for determining stocking rates, improving grazing efficiency, and supporting economic sustainability. Traditional methods to quantify forage mass, however, are time-consuming and labor-intensive. This thesis presents research efforts to achieve accurate forage mass estimation in cereal rye pasture by integrating novel sensing technologies.

Chapter 1 reviews the importance and evolution of forage mass estimation methods, from destructive clipping to indirect approaches based on canopy measurements. It then introduces remote sensing, from proximal tools to drone- and …

Llm-Driven Semantic Explanations For Soil Moisture Prediction Models, Bamory Ahmed Toru Koné, Khouloud Boukadi, Rima Grati, Emna Ben Abdallah, Massimo Mecella

All Works

Efficient soil moisture prediction is crucial for sustainable agricultural practices, especially in the face of climate change and increasing water scarcity. However, the adoption of machine learning (ML) models in this context is frequently limited by their lack of interpretability, particularly among non-expert users such as farmers. This study proposes a novel approach to soil moisture prediction that combines high predictive performance with enhanced explainability. We propose a framework that leverages large language models (LLMs) to generate textual explanations based on a proposed irrigation and soil moisture ontology, thus making the model's predictions more understandable to farmers. The ontology formalizes …

Applying Ni-Wc As A Composite Coating Combines Both Protective And Repair Functions By Thermal Spray For Hptb Applications, Alaa Dahesh, Ahmed Hassan, Hussain Yousif

Journal of Applied Sciences and Nanotechnology

This work presents a solution to repair the deteriorating High-Pressure Turbine Blades (HPTB) instead of costly replacement by implementing a two-coating layer system for protection and repair. The first layer is a Bonding Coating (BC) that is made from a nickel-based alloy (NiCrBSi); the second layer is a Top Coating (TC), which is composed of a nickel-based composite material that contains tungsten carbide (WC) particles. Both layers were applied to a nickel-based substrate using a flame thermal spraying (FTHS) technique, with the spray parameters chosen based on the Taguchi method. The results were analyzed using optical microscopy, scanning electron microscopy …

Personalized Risk Prediction For Cancer Survivors: A Generalized Bayesian Semi-Parametric Model Of Recurrent Events With Competing Outcomes, Nam Hoai Nguyen, Seung Jun Shin, Elissa Dodd-Eaton, Jing Ning, Wenyi Wang

Faculty, Staff and Student Publications

Multiple primary cancers are increasingly more frequent due to improved survival of cancer patients. Characteristics of the first primary cancer largely impact the risk of developing subsequent primary cancers. Hence, model-based risk characterization of cancer survivors that captures patient-specific variables is needed for healthcare policy making. We propose a Bayesian semi-parametric framework, where the occurrence processes of the competing cancer types follow independent non-homogeneous Poisson processes and adjust for covariates including the type and age at diagnosis of the first primary. Applying this framework to a historically collected cohort with families presenting a highly enriched history of multiple primary tumors …

Forecasting Chemoradiation Response Midtreatment For High-Grade Gliomas Through Patient-Specific Biology-Based Modeling, David A Hormuth, Maguy Farhat, Bikash Panthi, Holly Langshaw, Mihir D Shanker, Wasif Talpur, Sara Thrower, Jodi Goldman, Sophia Ty, Calliope Custer, Jeanne Kowalski, Thomas E Yankeelov, Caroline Chung

Faculty, Staff and Student Publications

Purpose: The entire course of radiation therapy (RT) for high-grade glioma (HGG) is currently derived from pre-RT magnetic resonance imaging (MRI). Although it is possible to adapt RT during the course of treatment, it is often guided only by anatomical changes to the tumor. This study seeks to determine if a biology-based mathematical model, parameterized by patient-specific, multiparametric MRI (mpMRI) data, can accurately forecast HGG response during RT.

Methods and materials: Twenty one patients with HGG planned for 6 weeks of concurrent RT and chemotherapy were imaged weekly with mpMRI during RT and at 1, 2, and 3 months post-RT. …

Normalization Of Temperature Effects For Quality Assurance Of Quantitative Prostate Apparent Diffusion Coefficient Imaging Across Multiple Sites, Ken-Pin Hwang, Joshua Yung, R Jason Stafford, Caroline Chung, Aradhana M Venkatesan

Faculty, Staff and Student Publications

Background: Apparent Diffusion Coefficient (ADC) as measured by diffusion weighted imaging is known to negatively correlate with prostate tumor aggressiveness. Heterogeneity in system and protocol performance causes potential variability in ADC acquired across a large scanner network, prompting a need to evaluate quantitative ADC from a prostate-specific MR diffusion protocol as part of quality assurance (QA). Due to the temperature dependence of ADC, repeatability and reproducibility assessments typically require phantoms to maintain a temperature of 0°C, imposing a considerable burden when assessing large numbers of scanners.

Purpose: To develop a QA procedure at room temperature for assessing the reproducibility of …

Impacts Of Halophila Stipulacea Invasion On Organic Matter Dynamics In Thalassia Testudinum Dominated Seagrass Meadows, Puerto Rico, Jonathan D. Herald

Master's Theses

Halophila stipulacea (Forsskål & Niebuhr) is a tropical seagrass native to the Red Sea, Persian Gulf, and Indian Ocean. In 2002, invasive populations of H. stipulacea were identified in the Caribbean Sea where it now forms large meadows, displaces native seagrasses, alters invertebrate and fish communities, and potentially affects organic matter dynamics in invaded meadows. This study investigated differences in organic matter dynamics between meadows of H. stipulacea and native Thalassia testudinum within the Jobos Bay National Estuarine Research Reserve (JBNERR) on the southern coast of Puerto Rico. From March 2023-March 2024, sampling was conducted across four sites to investigate …

Effects Of Urbanization On The Population Genetics And Morphology Of The White-Footed Mouse (Peromyscus Leucopus) Across An Urban–Rural Gradient, Sydney A. Morton

Master's Theses

Urbanization alters landscapes in ways that can reshape wildlife populations by changing habitat connectivity, resource availability, and selective pressures. To investigate how urbanization influences both genetic differentiation and morphology in small mammals, I examined Peromyscus leucopus (PELE) across an urban–rural gradient in northern Georgia. Using RADseq SNP data from PELE individuals across 13 sites, I quantified pairwise genetic differentiation (F_ST) and tested how various environmental variables, including impervious surface cover, human population density, and tree canopy cover, predict genetic differentiation after accounting for geographic distance. Genetic differentiation increased with geographic distance and, after controlling for isolation by …

The Interplay Between Structure And Teacher Agency In The Detracking Of Algebra 1: A Qualitative Case Study, Kristine Norris

Dissertations

This single qualitative case study examined how detracking Algebra 1 at Mountainview High School (pseudonym) restructured the course’s rules and resources and how teachers exercised agency in response. Guided by Giddens’ structuration theory and Weis and Fine’s critical bifocality, the study explored how institutional structures both shape and are reshaped by teacher action within broader systems of inequality. Participants were seven Algebra 1 teachers involved in the first two years of detracking. Data was gathered using semi-structured interviews and analysis focused on two research questions: (1) In what ways did detracking alter the structures of Algebra 1 courses, and (2) …

Teaching For Global Ecological Citizenship: A Case Study Of Rural Teachers’ Experiences And Practices, Sarah L. Tennyson

Dissertations

The purpose of this qualitative case study is to examine the experiences and practices of rural educators incorporating global ecological citizenship education into the social studies classroom. The systematic themes of social studies education in the past must evolve with the diversity of the classroom. Traditional themes taught must become more globally relevant, including ecological citizenship, the responsibility of citizens to sustain and participate in global ecological issues. One on one interviews with four rural Georgia educators along with artifacts they provided were examined for themes related to two research questions. How do rural educators adapt their instructional strategies to …

Occurrence Of Per And Polyfluoroalkyl Substances In Rural Wastewater Treatment Plants And Comparison To Large Treatments Plants, Madeleine Rauhauser

Department of Civil and Environmental Engineering: Dissertations, Theses, and Student Research

Per and polyfluoroalkyl substances (PFAS) are a class of compounds defined by their carbon – fluorine bonds, which are incredibly strong. The strength of this bond makes these compounds very useful industrially, they are used as waterproof and grease proof coatings, and as fire retardants. The same thing that makes them useful industrially makes them incredibly recalcitrant in the environment, it is difficult to destroy these compounds. PFAS are also suspected to cause health impacts in people, such as kidney cancer and low birth weights. Wastewater has been commonly identified as a contributor to the amount of PFAS in the …