Digital Commons Network^™

Towards Safe And Reliable Ai Models, Zihang Zou

Graduate Thesis and Dissertation post-2024

With the rapid advancement of artificial intelligence, its applications have become indispensable in our daily lives. Emerging advanced AI models, from general-purpose Large Language Models (LLMs) such as ChatGPT and Deepseek to domain-specific systems such as autopilot for autonomous vehicles, are changing the way we work and live by increasing efficiency and fostering innovation. However, these advancements also bring significant risks, as they may compromise data ownership or yield unreliable outcomes. In response, we focus on two critical areas: data safety and model reliability. To ensure data safety and promote lawful data processing, we investigate anti-neural watermarking techniques designed to …

Contagion Chronicles: A Spatiotemporal Analysis Of Armed Conflict And Infectious Disease, Devyn Escalanti

Graduate Thesis and Dissertation post-2024

This dissertation examines the relationship between armed conflict and infectious disease transmission through a novel theoretical framework that integrates ecological disturbance and health systems resilience theories into epidemiological contexts. The framework conceptualizes diseases along a continuum from "pulse" to "press." Pulse diseases, like Ebola, are characterized by sudden onset, high mortality, and short-term outbreaks. Press diseases, like HIV, progress gradually with lower mortality and long-term persistence. Grounded in this pulse–press disease typology and ecological and health systems resilience, the theory suggests that the absorptive capacity, adaptive capacity, and resilience of community health workers in conflict areas can influence disease transmission, …

Shock Tube Investigation And Chemical Kinetic Modeling Of Green Propellants For Space Missions, Marley Albright

Graduate Thesis and Dissertation post-2024

Hydrazine thrusters have been used for decades to land spacecraft on other planets. While its use is common, handling hydrazine is difficult due to its toxicity; this poses a risk to humans during testing and loading the propellant onto a spacecraft. Occupational Safety and Health Administration has limited human exposure to 1 part per million (ppm) of hydrazine per eight-hour workday. Green propellants are promising alternatives because they have lower toxicity but do not sacrifice performance. Green propellant technology has been demonstrated in space and future missions are planned, such as NASA’s Green-Propulson Dual-Mode technology demonstration. At the University of …

Authorship Attribution Of Ai-Generated Codes, Soohyeon Choi

Graduate Thesis and Dissertation post-2024

The field of code authorship attribution focuses on identifying the author(s) of source code written in a specific programming language. Various methods, including manual crafting, automated crafting, and deep learning-generated features, have been developed for this task. These techniques leverage the unique stylistic patterns inherent in code, such as structure, comments, variable names, and function names, to attribute authorship accurately. This field has practical applications in software forensics, cybersecurity, and code plagiarism detection, enabling investigations into issues like piracy, intellectual property violations, and malware attribution. In educational settings, it helps detect plagiarism in programming assignments. However, traditional methods face challenges …

Structure-Property Relationship For Silver-Incorporated Cerium Oxide Nanoparticles, Shanza Nusrat

Graduate Thesis and Dissertation post-2024

Cerium oxide nanoparticles are known for their applications across several domains due to their biocompatibility, high oxygen storage capacity, redox behavior, and stability. However, their intrinsic electrical conductivity is relatively low, which can limit their performance in energy storage applications. Doping cerium oxide with transition metals has proved to enhance conductive properties and, hence, their performance in related applications. This research investigates the doping of cerium oxide with silver (Ag) and its correlation with the properties of the material. Silver-doped cerium oxide nanoparticles were synthesized by wet-chemical synthesis at room temperature by a facile procedure. Cerium oxide was doped with …

Computation-Efficient And Scalable Robotic Motion Planning Techniques To Address Dynamic Environmental Constraints, Apan Dastider

Graduate Thesis and Dissertation post-2024

In modern robotics, effectively computing optimal robotic control policies under dynamically varying environmental constraints poses substantial challenges and remains a unique research endeavor. To compute efficient and safe robotic motion planning while achieving collective objectives, robotic agents have to satisfy two criteria. First, the working robotic entity must be capable of handling a non-stationary working environment with dynamic obstacles and system constraints. Second, to ensure the real-time response, the robotic agent has to compute an effective control policy that meets a real-time learning performance. Despite significant advancements in motion planning with the appearance of extensive computing resources and advanced deep …

Assessing Driving Performance And Safety In Managed Lanes: A Human Factors Study On The Effects Of Buffer Separation Treatments With Delineators, Sharfuddin Ahmed

Graduate Thesis and Dissertation post-2024

Buffer separation with delineators is one of the commonly used separation treatments across US highways to distinguish managed lanes (ML) from general purpose lanes (GPLs). As MLs continue to be integrated into US highway system for improving traffic flow and reducing congestion, this separation method has become more popular due to its flexibility and cost efficiency over more robust physical separations. While past studies have explored the effectiveness and safety of buffer separations with delineators compared to other treatments, none have examined how specific changes within the buffer design—like separation width and delineator height—impact driver behavior. It is particularly important …

Evaluating Pedestrian Behavior In Response To Innovative Detection Messages And Wait-Time Indicators At Signalized Intersections: A Virtual Reality Study, Nada K. Ahmed

Graduate Thesis and Dissertation post-2024

Pedestrian safety at signalized intersections remains a growing concern in urban environments, necessitating effective communication between pedestrians and traffic control devices. Traditional approaches often prove inadequate in addressing these challenges. This study utilized virtual reality simulation to examine how innovative detection systems and wait-time indicators affect pedestrian behavior under varying environmental conditions. Ninety participants evenly distributed across three age groups experienced 18 crossing scenarios that differed by detection type, wait indication, crossing time, crossing distance, visibility, traffic density, and time of day. Key behavioral measures including start-up lost time, crossing speed, and total crossing duration were analyzed using linear mixed-effects …

Development Of A Laser Spectroscopic Device For Measurements Of Temperature And Speciation In Detonation Environments, Andrew M. Derusha

Graduate Thesis and Dissertation post-2024

Characterizing novel detonation-based combustion systems is critical to assessing new designs and improving the understanding of their unique operations. Due to the harsh nature of a detonation environment, industry-standard sensing and diagnostic tools can fail to provide adequate methods for in-situ quantitative measurements of key thermodynamic properties. Diagnostic equipment must adapt to meet the needs of new combustion systems. Laser absorption spectroscopy (LAS) is a technique commonly utilized in combustion research to provide such measurements in laboratory settings due to its precision, minimally invasive properties, and high data acquisition rates. To prepare such a sensor for use in a challenging …

Gait Kinematics And Muscular Responses To Rhythmic Perturbations In Healthy Young Adults, Kayley A. Romero

Graduate Thesis and Dissertation post-2024

Falls pose a significant risk across all ages, leading to injury and loss of independence. Understanding gait adaptation to external perturbations provides insights into balance control and muscle activation, with implications for rehabilitation and fall prevention. While entrainment, the synchronization of biological rhythms to external stimuli, has been widely studied using auditory and visual cues, its effects in response to mechanical perturbations remain unclear. This study examines how young, healthy adults adapt to discrete mediolateral perturbations and whether entrainment occurs when mediolateral perturbations are applied at set timings determined by their natural stride period.

Participants walked on a self-paced treadmill …

On The Dynamic Connectivity And Phase Transition In River Networks, Juthika Roy

Graduate Thesis and Dissertation post-2024

River network (RN) structure has a significant impact on flux transport, aggregation and ecosystem connectivity. In this dissertation, the structural control of RN on flux transport was investigated by the application of integrated framework of dynamic clusters and network graph theory. This new framework has been proposed to determine sub-catchment connectivity using minimum and maximum flow criteria. Our analysis on synthetic and natural RNs across United States demonstrates that basin scale connectivity at low flow conditions is efficiently governed by the abundance of side branching junctions whereas this characteristic slows down the flux convergence rate at maximum flow condition.

Furthermore, …

E-Cskpd: A Frequentist Framework For Multi-Modal High-Order Tensor Regression, Joshua D. White

Graduate Thesis and Dissertation post-2024

Understanding which features drive the outputs of modern learning systems remains a core challenge for regression and classification. While deep neural networks can be highly accurate, their latent representations often obscure how inputs map to decisions, limiting statistical inference, stakeholder trust, and practical use in domain decision-making. Sparse Kronecker Product Decomposition (SKPD) is a frequentist, tensor-structured approach that offers interpretable regression/classification in high dimensions. However, standard SKPD restricts relationships among predictors and covariates and is typically limited to order-2 tensors. This dissertation introduces E-CSKPD, an ensemble framework that (i) revisualizes designs to control non-imaging covariates; (ii) applies invertible, CNN-inspired transforms …

Practical Applications For Biomechanical Modeling Methods, Joseph M. Dranetz

Graduate Thesis and Dissertation post-2024

Models are tools humans use to understand and assess the world around them. The field of biomechanics is one well suited to modeling pursuits. Physiological systems are complex, and modeling may serve as an appropriate method in understanding them. Recently, technologies enabling the collection of biomechanical data have improved greatly and continue to expand. This dissertation serves as an anthology of various attempts to collect and interpret biomechanical data with modeling.

The first study: Electromyographic, ultrasound, and knee torque data was collected from both legs of an individual with a unilateral transtibial amputation. These data were used to generate a …

Strengthening Community Resilience Towards Infrastructure Disruptions Due To Hurricanes, Tasnuba Binte Jamal

Graduate Thesis and Dissertation post-2024

Extreme events such as hurricanes can significantly disrupt infrastructures, leading to negative impacts on productivity, economy, and social wellbeing. Understanding community resilience against infrastructure disruptions in response to such events is crucial for effective preparedness, mitigation, and recovery efforts. The overarching problem addressed in this dissertation centers around enhancing community resilience against infrastructure disruptions due to hurricanes. Community resilience is defined as the ability of a community to return to normalcy following a disaster. One way to enhance community resilience is to strengthen its critical lifelines and infrastructure services (e.g., power, water, transportation, internet) and capital—to reduce the likelihood of …

Inter-Basin Groundwater Flow In West-Central Florida, Sally Yousef Elrashedy Dr

Graduate Thesis and Dissertation post-2024

The mean annual values of inter-basin groundwater flow (IGF) and its inter-annual and seasonal variability across 172 watersheds in west-central Florida were estimated using the Integrated Northern Tampa Bay (INTB) model, which is an application of the Integrated Hydrologic Model dynamically coupling HSPF and MODFLOW. The estimated IGF for the scenario without human fluxes shows spatial heterogeneity in the model domain ranging from -1291mm/year to 4808 mm/year, and the variability of IGF decreases with spatial scales. A region with gaining IGF is defined as groundwater importer and a region with losing IGF is defined as a groundwater exporter. The characteristics …

The Effects Of Camera, Environmental, And Motion Properties On Simultaneous Localization And Mapping, Esteban Segarra Martinez

Graduate Thesis and Dissertation post-2024

Simultaneous localization and mapping (SLAM) is the process of estimating the pose of a robot within an unknown environment. A variant, visual SLAM (V-SLAM), estimates the pose only using visual information from a camera. In recent years, the importance of V-SLAM has increased and expanded to applications such as robotics, virtual reality, augmented reality, and autonomous driving. Accurate estimation of poses is critical as these applications need to quickly perceive an environment and update its pose under complex scenarios. V-SLAM is composed of algorithms that utilize robot vision to distinguish landmarks in the environment in order to estimate the geometric …

Integration Of System Dynamics And Agent-Based Simulation To Emulate Cyberattacks In An Iot-Based Smart Grid, Francisco Valdez

Graduate Thesis and Dissertation post-2024

The electric power grid is a cyber-physical system (CPS) that plays a fundamental role in modern society. With the integration of renewable energy sources and advanced communication technologies, Smart Grids (SGs) can enhance both the profitability and reliability of the electric power system. The communication network that interconnects numerous remotely distributed generators, devices, and controllers plays a vital role in grid control, and current trends favor the widespread adoption of Internet of Things (IoT) devices. However, this network is inherently vulnerable to cyberattacks. This dissertation presents a hybrid methodology to model and analyze the dynamic behavior of an electrical Smart …

Critical State-Based Study On Deformations Due To Construction Activities And Blast Densification, Jorge Enrique Orozco-Herrera

Graduate Thesis and Dissertation post-2024

The behavior of geo-structures and foundation systems in densely populated urban areas relies on the successful performance of the supporting soils when subjected to static (e.g., the weight of the structure or service loads) or dynamic (e.g., earthquake-induced ground motions, construction-induced vibrations, wind loads, or traffic) loading scenarios. Construction-related activities and ground improvement operations can also be responsible to alter the stress-strain-strength behavior of supporting soils by causing vibration-induced ground deformations, excessive vibrations, or mobilizing soil shear strength in relation to in situ or as-built conditions. Construction-related activities and ground improvement operations such as blast densification trigger complex dynamic input …

Additively Manufactured Engineered Transparent Ceramics For Laser Gain Media, Ross Osborne

Graduate Thesis and Dissertation post-2024

Transparent ceramics (TCs) offer a key advantage over single crystals as solid-state laser materials, in that their composition can be spatially varied to improve laser performance. In this work, additive manufacturing (AM) techniques were applied to TC fabrication to produce laser gain media with engineered gain and index profiles. We have used direct ink write (DIW) to fabricate laser rods and channel waveguides and introduced ink-spraying, a new technique, to realize multilayer planar waveguides. Modifications to the DIW process, including the use of sacrificial drying layers to promote uniform drying, the optimization of the drying, burnout and calcination steps all …

Fundamental Mechanisms For Flame Acceleration And Detonation Phenomena, Rachel Hytovick

Graduate Thesis and Dissertation post-2024

Hypersonic propulsion via scramjets and detonation engines is enabled through the acute understanding of deflagrations, detonations, and multiphase interactions. Deflagrations are explored to understand the compressible flame regime for scramjet unstart and detonation transition in reacting systems. Compressible flames are experimentally classified with a comprehensive diagnostic suite combining particle image velocimetry (PIV), chemiluminescence, schlieren, and pressure measurements in a turbulent shock tube. Several gaseous fuels are explored to identify the universal mechanism responsible for flame compressibility and acceleration. The ability to compress deflagrations and accelerate the system to a detonation directly relies in the successful conversion of energy available to …

A Comparative Analysis Of Figurines From The Lower Rio Verde Valley From Oaxaca, Raynaliz C. Velazquez

Graduate Thesis and Dissertation post-2024

This thesis investigates 618 ceramic figurines from the Lower Río Verde Valley in Oaxaca, Mexico, to explore how figurine use varied across civic-ceremonial, domestic, and outlying contexts during the Terminal Formative and Late Classic periods. The central hypothesis proposes that figurines from the Río Viejo Acropolis, a civic-ceremonial center, would be larger and exhibit greater labor investment than those from residential and peripheral sites, reflecting more labor-intensive ideologies and ceremonial practices. Through typological, spatial, and material analysis, the study examines attributes such as size, surface treatment, production method, and the presence of musical elements. The dataset includes newly analyzed figurines …

Bim-Based Machine Learning Surrogate Models For Energy And Carbon Prediction To Support Leed Certification Evaluation, Liliane Magnavaca De Paula

Graduate Thesis and Dissertation post-2024

This research developed and validated a Building Information Modeling (BIM)-based Machine Learning (ML) framework for predicting the energy and carbon performance of office buildings during early design stages. The study addresses the need for reproducible data-driven methods supporting sustainable design decisions with reduced simulations time. The proposed approach integrates Revit and Insight with statistical modeling in Weka, creating an automated, transparent, and regionally adaptable workflow for energy and carbon prediction from a BIM-generated data. A reduced-factorial Design of Experiments (DOE) guided the generation of 260 parametric Insight simulations, including base, generalization, and stress-test models distributed across six U.S. climate zones. …

Exploring The Experience Of Heart Disease Among Burmese Immigrants In The United States, Khup S. It

Graduate Thesis and Dissertation post-2024

The ongoing political and military turmoil in Myanmar has resulted in a significant increase in the migration of Burmese individuals to the United States in recent years. This demographic shift has brought attention to the health challenges faced by Burmese immigrants, particularly with regard to heart disease (heart attack). The complexities of the U.S. healthcare system, compounded by legal and social barriers, place immigrants at greater risk for health disparities. This study examines the health challenges of Burmese immigrants in Tulsa, Oklahoma, with a specific focus on their perceptions and experiences related to heart disease, as explored through illness narratives. …

An Integrated Approach To Waste Management And Development Of Data Driven Decision Support Tools, Mert Gokgoz

Graduate Thesis and Dissertation post-2024

Municipal solid waste management requires balancing environmental performance, operational efficiency, and community participation. Proper recycling contamination in single-family units is a persistent barrier to material recovery, reducing the effectiveness of local programs by increasing recycling contamination. This dissertation develops an integrated, data-driven framework that combines behavioral modeling of communities via statistical analysis and machine learning (ML), and life cycle assessment (LCA) to support decision-making in residential waste management. The first component examines household recycling behavior using data from Orange County, Florida’s Recycling Quality Improvement Program (RQIP). Cart-tagging results were integrated with socio-demographic data from the U.S. Census to predict contamination …

Shock Tube Droplet Breakup Experiments Facilitated By Acoustic Levitation, Anthony J. Aguilera

Graduate Thesis and Dissertation post-2024

Droplet introduction methodology in a shock tube differs based on the experimental conditions. For experiments in which the objective is to image droplet breakup, acoustic levitation allows for a consistent droplet shape and for diagnostic equipment to be aligned to a droplet prior to an experiment. A levitator was designed and used in the University of Central Florida’s shock tube facility. Polystyrene doped droplets were levitated to increase the likelihood of cavitation occurring. Multiple droplets were impacted by a shock wave, and the breakup regime was captured using a high-speed camera.

Robust Low-Rank Tensor Discriminant Analysis With Optimal Scoring, Kyle Brinker

Graduate Thesis and Dissertation post-2024

Robust discriminant analysis with optimal scoring has been shown effective for classifying high-dimensional matrix data, particularly in the context of two-dimensional images. The proposed method presents an extension of the Multi-Projection Optimal Scoring Discriminant Analysis framework introduced by Huang \& Zhang (2020), incorporating low-rank Tucker decomposition and regularization techniques to address higher-dimensional cases. The enhanced method is designed to effectively handle tensor-structured data, which is prevalent in imaging applications. The model creates a sparse discriminant projection tensor B that captures class-specific features, such as ball-like patterns in image data and indicators of Alzheimer’s disease in Magnetic Resonance Imaging (MRI) scans. …

Optimizing Emissions And Sustainability Of Alternative Fuel Buses Using Machine Learning And Life-Cycle Modeling, Md Rezwan Hossain

Graduate Thesis and Dissertation post-2024

Public transportation is undergoing a critical transition as agencies seek to balance environmental responsibility, operational efficiency, and economic viability. Alternative fuel buses (AFBs), including electric, compressed natural gas (CNG), propane, and biodiesel, offer opportunities to reduce emissions and advance sustainability, yet their adoption raises complex trade-offs. High upfront costs, infrastructure constraints, range limitations, and variable real-world performance complicate decisions for fleet managers and policymakers. This dissertation addresses these challenges by integrating real-world operational data, life-cycle modeling, and machine learning within a multi-objective optimization framework to evaluate and optimize AFB deployment. Using detailed telematics and GPS datasets from Central Florida's transit …

Design Tool For Ammonia Cracker Hardware, Andrew Menendez

Graduate Thesis and Dissertation post-2024

Ammonia is a promising carbon-free fuel for commercial aviation and represents a viable pathway toward net-zero emissions. However, direct ammonia combustion is inefficient, creating the need for onboard cracking to generate hydrogen for improved combustion characteristics. One pathway for onboard cracking is to have the hardware live around the combustor liner where heat generated from the combustion process can be used to sustain the reaction. This study aims to develop a design tool for the ammonia cracking reactor hardware. A simplified, single-step surface reaction mechanism was implemented using ANSYS FLUENT and CHEMKIN PRO R1, demonstrating strong agreement with experimental data …

Smart Urban Air Pollution Management Using Crowdsourced Big Data, Low-Cost Sensor Network And Aerial Remote Sensing, Md Hasibul Hasan

Graduate Thesis and Dissertation post-2024

Air pollution remains one of the most pressing environmental and public health challenges, especially in urban areas with dense populations and concentrated emission sources. Effective air quality management requires accurate information on emissions to guide control strategies and reliable exposure estimates to evaluate health impacts. This dissertation examines multiple sources of uncertainty in both emission and exposure assessments. Two major emission sectors were investigated: on-road motor vehicles and landfills. For on-road sources, detailed traffic information inferred from crowdsourced big data were leveraged to improve the estimation of traffic-related emissions. For landfills, the application of low-cost sensor and aerial remote sensing …

Towards Achieving General Video Understanding, Rohit Gupta

Graduate Thesis and Dissertation post-2024

This dissertation investigates the problem of general video understanding, aiming to develop models capable of robust, adaptable, and semantically rich interpretation of video data across diverse domains and tasks. Despite significant progress in video analysis, existing methods remain limited by restricted vocabularies, inadequate fine-grained discrimination, and vulnerability to noise and adversarial perturbations. To address these challenges, the research advances four complementary directions encompassing multimodal fusion, open-vocabulary recognition, robustness enhancement, and unified video–language modeling. First, a multi-modal prototype contrastive framework is proposed for fine-grained video classification, integrating visual and auditory cues through a transformer-based architecture and supervised contrastive learning with class …