Physical Sciences and Mathematics Commons^™

Advancing Data Usability, Activity Modeling, And Stability Optimization In Computational Enzyme Design, Spencer Gardiner

Theses and Dissertations

A grand challenge of computational biology is to computationally design, in a single pass, a protein sequence that catalyzes an arbitrary chemical reaction at a high rate under specified conditions [1]. This work details advances in three essential areas on the path to that goal: data quality, activity prediction and modeling, and stability optimization. The structure and implementation of the Allotrope Simple Model (a FAIR data format for many scientific instruments) was examined in [2], setting the stage for training deep learning models on high-quality experimental datasets from diverse sources. In [3], the limits of physics-based and deep learning tools …

Beyond Single Images: A Comprehensive Benchmark For Album-Level Vision-Language Understanding, Shawn Huang

Theses and Dissertations

Automatic album organization has been studied extensively over the past decades due to significant progress in digital photography. Recent Vision-Language Models (VLMs) have shown strong performance on multi-image understanding, making them natural candidates for automating album organization workflows. While VLMs’ abilities in multi-image understanding have been widely studied, their performance on album organization remains underexplored. To bridge this gap, we introduce AlbumBench, the first comprehensive benchmark for automatic album organization. Specifically, we (1) define album organization tasks as photo selection for album-specific user objectives, photo rating according to how well user intents are fulfilled, and album-specific photo grouping given a …

Tuning Ion Mobility And Molecular Confinement In High-Performance Polymer Electrolytes For Energy Storage, Ezzeldien Yousef Muhammed Yousef

Theses and Dissertations

This research addresses the critical energy density limitations of aqueous supercapacitors, which are traditionally constrained by the narrow electrochemical stability window (ESW) of water 1.23 V. By employing two distinct molecular engineering strategies, this study developed high-performance electrolyte systems that significantly extend voltage stability and thermal resilience.

The first system, CsBr@PAM/HA, utilizes a polyacrylamide and hyaluronic acid hydrogel matrix. This system exploits the chaotropic nature of Cs⁺ ions to disrupt the aqueous hydrogen-bonding network, enhancing ionic conductivity to 104 mS cm^-1. Through systematic salt screening, CsBr was identified as the optimal electrolyte, enabling a stable 2.0 V …

Engineering Nanoelectrocatalytic Systems For Co2 And Nitrate Conversion Into Value-Added Chemicals, Abdelrahman Mohamed Abdelmohsen

Theses and Dissertations

The thesis addresses two major related environmental crises: nitrate pollution of water systems and the ever-increasing levels of atmospheric carbon dioxide. Both challenges are closely linked to anthropogenic disturbance of nitrogen and carbon cycles and require sustainable, energy efficient mitigation techniques. In this study, we investigate electrochemical conversion routes as a unified approach to convert these pollutants into value-added compounds: ammonia (NH3) via nitrate reduction (NO3-RR) and ethylene (C2H4) via carbon dioxide reduction (CO2RR). The first half of this work deals with the design and engineering of Cu-Zn alloy electrocatalysts for efficient NO3-RR. Tuning the alloy composition and surface nanostructure …

Design Principle For Staple-Mediated Stabilization Of Peptides And Proteins, Samantha C. Hatfield

Theses and Dissertations

Macrocyclization, or stapling, has emerged as a powerful strategy to enhance the stability and function of peptides and proteins by constraining their conformational ensembles. Stapling can improve thermodynamic stability, increase resistance to proteolytic degradation, and enhance biomolecular recognition, making it particularly valuable for the development of peptide-based therapeutics. Despite these advantages, staple design remains largely empirical, relying on iterative screening of linker chemistries, lengths, and attachment sites. This trial-and-error approach limits scalability and hinders the rational design of stabilized peptide systems. In this work, we address this challenge by integrating large-scale experimental datasets with molecular dynamic simulations to develop predictive …

Beyond Emergency: Charting Pathways To Sustainable Development In Fragile And Conflict-Affected Settings: A Case Study Of Yemen, Sharif Abdurabu Salem Farid Alsahbool

Theses and Dissertations

For nearly a decade, Yemen's eastern governorates of Hadramawt, Shabwah, and Al‑Mahrah have been relatively stable, yet the international aid system has kept delivering short‑term humanitarian relief like food baskets and water trucking instead of shifting to sustainable development that could rebuild people's lives, create jobs, fix infrastructure, and strengthen local government. This thesis investigates why this transition has taken so long. Based on 21 interviews with government officials, international NGO staff, and local NGO managers in these three governorates, the study develops a new way of understanding the problem. It introduces three ideas: the 'emergency mentality', which is a …

Context-Sensitive Control-Flow Analysis For Non-Standard Language Semantics, Tim Whiting

Theses and Dissertations

Control-Flow Analysis (CFA) is a static technique for understanding the execution of higher-order programs without running the code. CFA aids in identifying bugs, assisting program comprehension, and exposing optimization opportunities. Constructing a CFA typically begins with an idealized evaluator--such as an abstract machine or interpreter--and proceeds by approximating components of evaluation to ensure that: (a) the analysis is sound (representing all possible behaviors), and (b) the analysis is guaranteed to terminate (finitely representable). These constraints necessitate a loss of precision, providing an approximation of the values that flow to program locations rather than exact execution traces. This imprecision causes the …

Data Science In The Utah Draft K-12 Standards Of Mathematics, Kate Webster Green

Theses and Dissertations

In today's data-driven world, the field of data science is becoming increasingly important. Research has shown that more job listings in Utah require applicants to have data science skills than ever before. Due to this growing demand, there is a push to begin data science instruction in grades K-12; however, what data science instruction is in grades K-12 is not yet clearly defined. Previous research has identified the skills and processes that make up data science in grades K-12, and several states and research groups have begun developing and implementing data science standards and learning progressions. This study analyzes the …

Scientific Crosstalk: Natural Language Processing, Praveshika Bhandari

Theses and Dissertations

While sentiment analysis has made significant strides in domains such as social media and personal correspondence, its application to formal scientific writings remains under-explored. The crosstalk between emotional expressions in personal and professional communications has also received limited attention despite its potential to reveal insights into the emotional drivers of scientific creativity. Our research introduces a computational framework designed to detect and quantify emotional expressions across various documents over time. Leveraging state-of-the-art transformer models fine-tuned on domain-specific corpora, the framework models emotional tone distribution. Integrating emotion analysis with knowledge graph modeling enables the exploration of emotional trends alongside key scientific …

Spatiospectral Lobes And Radiation Modeling Of Partially Coherent Supersonic Jet Noise, Tyce Wayne Olaveson

Theses and Dissertations

Noise fields generated by high-performance military jets are known to contain acoustic structures that differ from their simulated and lab-scale counterparts. These differences may limit current noise control efforts, which are designed and tested at scale before being considered for full-scale aircraft. One such noise structure is the presence of spatiospectral lobes, which appear as multiple spectral peaks at single microphone locations and multiple radiation lobes in field reconstructions. This dissertation uses data collected from the T-7A jet noise measurement to characterize the spatial behavior of the spatiospectral lobes as well as their first time-domain description. A multiple wavepacket decomposition …

Traveling Wave Fronts For Sir Epidemic Models With Nonlocal Dispersal And Delayed Effects, Ashley Evette Embry

Theses and Dissertations

This dissertation investigates traveling wave solutions for two classes of delayed nonlocal dispersal susceptible--infected--recovered (SIR) epidemic models incorporating biologically realistic mechanisms such as delayed infectivity, delayed dispersal, nonlocal transmission, demographic turnover, and nonlinear incidence effects. These models extend classical spatial epidemic frameworks by allowing long-range population movement through nonlocal dispersal operators and incorporating temporal memory into both diffusion and transmission processes. The primary objective is to establish the existence of traveling wave solutions connecting disease-free equilibria to endemic states and to characterize threshold conditions governing epidemic propagation. The simultaneous presence of nonlocal dispersal, multiple delays, and non-monotone nonlinear incidence terms …

Vigor: Virtual Intelligence For Gaze Observation And Representation, Meherun Nesa Shraboni

Theses and Dissertations

Eye tracking in modern XR head-mounted displays can capture high-frequency gaze data at 90–120 Hz, generating thousands of samples within a single 30-second interaction. At the same time, the XR market has grown to millions of active users worldwide, with major platforms such as Meta investing heavily in eye-tracking-enabled devices. Despite this large-scale adoption and data availability, most applications still rely on hand-controller input, with gaze either processed offline or reduced to a simple pointing signal. As a result, the majority of temporally rich gaze information remains unused in real-time interaction, limiting system responsiveness, reducing interaction fidelity, and preventing effective …

Real-Time Instruction Internalization For Large Language Models, Brenden Smith

Theses and Dissertations

For the end user, Large Language Models (LLMs) are programs that process natural language inputs into natural language outputs. In popular usage, this tends to take the form of conversation: a user asks a question, provides information, or gives instructions, and the LLM (hopefully) replies in a manner we would expect of an informed and compliant person. While convenient and intuitive for users, this natural conversational format encourages the misconception that LLMs are learning from conversations, when they do not. This work presents the benefits and practicality of a language model paradigm that meets this user expectation -- that is, …

Adaptive Channel Switching For Contention Resolution, Shafqat Hasan

Theses and Dissertations

Contention resolution is a fundamental problem in distributed computing, where multiple devices compete to transmit over a shared channel without centralized coordination. Classical models typically assume a single always-available channel and focus on minimizing makespan. However, modern wireless systems increasingly operate under spectrum-sharing frameworks in which access to high-capacity spectrum is opportunistic and may be interrupted by higher-priority users. These settings introduce new challenges, including asymmetric channel speeds, adversarially scheduled evictions, and non-trivial switching costs. In this thesis, we study contention resolution in a dual-channel model consisting of a slow, always-available channel and a faster channel subject to adversarially scheduled …

Accelerating Defensive Cyber Operations Via Unsupervised Log Clustering And Automated Regex Template Synthesis, Charles Matthew Jones

Theses and Dissertations

Modern Security Operations Centers (SOCs) ingest millions of log entries per day, but manual parsing does not scale to the volume, heterogeneity, and rapid evolution of log formats. This dissertation investigates whether unsupervised clustering can automate the generation of candidate field-extraction templates while remaining deployable in production and feasible under realistic runtime and memory constraints. The central research question asks whether a machine-learning-assisted pipeline that proposes extraction templates for security engineer review—relative to reproducible human-authored baselines such as hand-written regular expressions—can measurably improve rule-set deployability and corpus-scale extraction quality. Improvement is quantified using four applicability metrics: Coverage Rate (CR), Exclusive …

Parhsom: A Novel Parallel Hierarchical Self-Organizing Map Implementation, Rebekah E. Lane

Theses and Dissertations

The digital age has completely transformed the way that information is processed and stored, which makes cybersecurity a crucial field of research. Cybersecurity contains many different domains, but this work focuses on Intrusion Detection Systems (IDSs). Within the literature, Hierarchical Self-Organizing Maps (HSOMs) have been used to create trustworthy, explainable, and AI-based IDSs. However, HSOMs are trained sequentially, which means that training HSOMs on large datasets is slow. This work presents a novel parallel HSOM architecture, called parHSOM. The purpose of this research is to investigate the effect that parallel computation has on the HSOM training time. parHSOM is tested …

Using Seafloor Geomorphology, Classification Schema, And Spatial Analysis To Predict Geologic Structures, With Applications For Mars, Emma Morgan

Theses and Dissertations

This study evaluated whether seafloor geomorphology can be used to predict subsurface fault locations along the Cascadia continental margin and explored whether such relationships can be transferred to Mars. Geomorphological classification schemas such as geomorphons, bathymetric position index (BPI), slope, rugosity, aspect, and curvature were derived from seafloor bathymetry data and used as predictor variables in maximum entropy (MaxEnt) models developed separately for Pre-Pleistocene, Pleistocene, Quaternary, and active fault datasets. Model performance ranged from AUC = 0.62 to 0.67, showing moderate predictive skill, with younger fault systems exhibiting stronger geomorphic associations. A secondary analysis showed a weaker but detectable relationship …

Evaluating The Spatial Relationship Between Geomorphology And Subsurface Structures On The Us Atlantic Margin With Applications To Other Terrestrial Planets, Andrea Caitlyn Stiles

Theses and Dissertations

High-resolution digital elevation models (DEMs) are widely available for terrestrial planetary bodies, but subsurface structural data remain limited. This research evaluated whether surface geomorphology could predict subsurface features, specifically faults and methane seeps, along the US Atlantic continental margin as an analog site. Faults were mapped with seismic interpretation and compared with quantitative geomorphic classifications derived from 100-m bathymetry including slope, aspect, rugosity, bathymetric position index, and geomorphon classes. Maximum-entropy modelling was used to predict fault occurrences from surface data. Results indicated that specific geomorphic landforms and slopes had the strongest control on fault presence, reflected in topographic expression of …

Assessing The Relationship Between Subsurface Geology And Surficial Geomorphology: Remotely Predicting Geologic Features And Geohazards Using An Elevation-Trained Machine Learning Algorithm In The Northern Gulf Of Mexico, Allison L. Wing

Theses and Dissertations

This thesis evaluates the capacity to predict subsurface geologic features and submarine landslide susceptibility using surficial geomorphology derived from bathymetric elevation data in the Northern Gulf of Mexico. Quantitative geomorphic variables including slope, curvature, aspect, rugosity, geomorphons, and Bathymetric Position Index were generated from 30-meter digital elevation models and used as explanatory variables in presence-only Maximum Entropy models. Known locations of faults, pockmarks, mud volcanoes, hydrocarbon seeps, and landslides (particularly intact scarps) were used to train and validate predictive models through k-fold cross validation. Model performance was assessed using omission rates and AUC values. Results demonstrate that specific geomorphic signatures, …

Beyond English: Auditing And Mitigating Cross-Lingual Data Contamination In Multimodal Large Language Models, Pavan Dharma Adapa

Theses and Dissertations

This thesis extends data contamination auditing for multimodal large language models to multilingual settings. Using LLaVA 1.5 and a high-fidelity French parallel dataset derived from ScienceQA, the study evaluates how performance changes when identical image-question pairs are translated from English to French. The resultsshow a substantial cross-lingual performance decline and frequent flips from correct English predictions to incorrect French predictions, indicating that benchmark performance can depend heavily on memorized English-specific patterns rather than stable multimodal reasoning. To address this weakness, the thesis introduces an inference-time mitigation strategy based on perturbation ensembling and cross-lingual consistency aggregation. The proposed method reduces instance-level …

Regional Variability In Tornado Tracks: A Gis-Based Study Of Surface Roughness And Land Cover Differences Between The Lower Midwest And Southeast United States, Joshua Morgan Dison

Theses and Dissertations

Throughout the years, much research has been dedicated to the physical processes leading to tornadogenesis and tornado decay. Much of this research is centered around the inner workings of the supercell itself. Recent studies are now showing that tornadogenesis and decay processes may be generated by interactions with the Earth’s surface. This study examines longtrack tornadoes from 2001-2016 in the southeast and lower Midwest regions of the United States. Utilizing the National Land Cover Database (NLCD) and American Meteorological Society / Environmental Protection Agency / Regulatory Model (AERsurface) surface roughness values, efforts are made to see if statistical significance is …

A Pedagogically Effective Conceptual Framework For The Resilience Of Unit Test Suites To Refactoring, Daniel Paul Knight

Theses and Dissertations

Unit test suites are intended to support safe and efficient source code refactoring, yet in practice they can hinder rather than help when tests are tightly coupled to implementation details. Such non-resilient tests require frequent co-evolution, consume valuable engineering time, and may disincentivize beneficial code improvements. While concepts such as passive and active resilience, test smells, and technical debt have been studied individually, they have not been integrated into a single actionable framework, nor has their pedagogical value been systematically evaluated. This dissertation introduces a novel conceptual framework for the resilience of unit test suites to refactoring, grounded in resilience …

Low-Dimensional Metal Halides For Light-Emitting Diodes (Leds): Photophysical Aspects And Device Engineering, Udara Madushanka Kuruppu Kuruppu Achchige Don

Theses and Dissertations

In recent years, low-dimensional metal halides have garnered significant interest due to their exceptional optical properties. This dissertation focuses on the structural, optical, and photophysical properties of both lead-based and lead-free low-dimensional metal halides and investigates their applications in light-emitting diodes. In Chapter II, the nature of Mn2+ doping in two-dimensional (2D) lead halide perovskites is discussed. 2D perovskites show strong excitonic and dielectric confinement due to their quantum well structures. Introducing optically active Mn²⁺ dopant ions into these 2D perovskites further enhances their optical properties and promotes energy transfer from the host to the Mn²⁺ dopants, resulting in dual …

Assessment Of Local Ecological Knowledge, Oyster Predation, And Habitat Function To Improve Management Of Nearshore Habitats, Cynthia Evelyn Marie Lupton

Theses and Dissertations

Coastal ecosystems along the northern Gulf of Mexico are ecologically and economically vital yet increasingly threatened by environmental change, habitat degradation, and knowledge gaps that complicate restoration and management efforts. This dissertation integrates social and ecological approaches to improve understanding of predator dynamics, stakeholder perceptions, and habitat function related to nearshore habitats. Thereby, providing information that can inform adaptive management and restoration in the region through addressing three overarching objectives: 1) evaluate stakeholder perceptions of oyster (Crassostrea virginica) predation and management practices; 2) quantify environmental and habitat factors influencing oyster drill (Stramonita haemastoma complex) abundance; and, 3) assess the influence …

Unbounded Derivations On Algebras Associated With Monothetic Groups: An Expository Thesis, Britton Phillips

Theses and Dissertations

This thesis is an expository exploration of the paper Unbounded Derivations in Algebras Associated with Monothetic Groups. Monothetic groups will be utilized to create a minimal system that gives rise to two C*-algebras, and once we have these algebras, unbounded derivations will be able to be defined on them. These derivations are able to be classified and decomposed into ”special” derivations, and these decompositions help simplify the comparisons between derivations on different algebras. In particular, we emphasize the classification, covariance properties, innerness and approximate innerness, and the lifting behavior of these derivations.

From Natural Language To Cryptographic Protocol Specifications: Evaluating Llms For Cpsa Generation, Martin Duclos

Theses and Dissertations

Formal verification can prove the security properties of cryptographic protocols, but translating natural language specifications into formal models requires specialized expertise, limiting the broader adoption of formal verification methods. This dissertation investigates whether large language models (LLMs) can lower this barrier by automatically generating Cryptographic Protocol Shapes Analyzer (CPSA) models from natural language protocol specifications. We evaluate three complementary strategies for improving LLM-based CPSA code generation through systematic experimentation across 104 protocols and 15 language models. First, we analyze prompt engineering and find that moderate structured guidance yields the most accurate outputs, while excessive prompt complexity degrades performance. Second, we …

Evaluation And Mitigation Of Bias And Toxicity In Open-Source Large Language Models Using Crows-Pairs And Bold, Sai Harika Gade

Theses and Dissertations

This thesis evaluates bias and harmful language generation in five open-source language models and tests practical mitigation methods that do not require retraining. Two masked models are assessed with a sentence-pair benchmark for stereotype preference, and three generative models are assessed with a prompt-based benchmark for harmful continuations across demographic domains. The study uses a unified experimental workflow to compare model behavior, summarize differences across bias categories, and measure changes after intervention. Results show that the masked models favor stereotypical content above a random baseline, while the generative models usually produce low average toxicity but still show uneven risk across …

Integrating Augmented Reality Visualizations Into Data Science Notebooks Using Microsoft Hololens 2, Derek Willis

Theses and Dissertations

Data-science notebooks support iterative analysis but are limited to two-dimensional (2D) displays. This work presents an approach to extend such environments with rapid augmented reality (AR) visualization while preserving conventional 2D workflows. An opensource R package was developed to convert notebook objects into three-dimensional (3D) models, export them in the Graphics Language Transmission Format (glTF), and transfer directly to a Microsoft HoloLens 2 via a USB connection for viewing in the native 3D Viewer application. The proposed workflow eliminates manual conversion and transfer steps required by earlier methods. A user study employing a post-session questionnaire indicated that participants found the …

Agentic Synthetic Data Generation For Automated Model Development, Frederick Eugene Diehl

Theses and Dissertations

Large Language Model research has made large strides in capabilities from sentiment analysis to writing code. These advancements have been realized thanks to research into specific capabilities such as prompting techniques. Language models today have demonstrated the ability to create content, transform, and classify. These capabilities are not limited to academic exploration but also found in commercial products that are positioning themselves from application augmentation to personal assistants. These commercial products tend to steer towards single actions such as “summarize this article” or “write a function that performs action...” In parallel research has continued to advance towards more advanced constructs …

A Governance-Aware Multi-Agent Framework For Enhancing Fairness & Temporal Accuracy In Disaster Response Systems, Md. Ashfaqur Rahman

Theses and Dissertations

Large Language Models (LLMs) have demonstrated significant potential in disaster-response decision support, however, their deployment in high-stakes humanitarian settings raises critical concerns regarding factual reliability, fairness, temporal validity, and governance compliance. Hallucinated outputs, demographic bias, and outdated recommendations can directly impact vulnerable populations and undermine public trust. This dissertation proposes a governance-aware multi-agent framework designed to enhance fairness and temporal accuracy in disaster-response systems through structured Retrieval-Augmented Generation (RAG), verification-driven orchestration, and adaptive correction mechanisms.The proposed architecture decomposes response generation into specialized agents responsible for real-time retrieval, fact-checking, bias auditing, temporal validation, threshold-based correction, and monitoring. By embedding governance constraints …