Digital Commons Network^™

Paleoethnobotany And Starch Grain Residue Analysis Collected From Ceramics At The White Marl Site, Jamaica, Lydia E. Kiernicki

Graduate Thesis and Dissertation post-2024

The White Marl site represents the largest known Taíno settlement and archaeological site in Jamaica. Road construction in the eighteenth century first disturbed the site, followed by further damage during Mandela Highway construction in the 1940s. The site now faces renewed threats from a proposed highway expansion that has divided it into two zones. In response, the Jamaica National Heritage Trust began archaeological investigations in 2016 to assess and document the land before development and preserve any recoverable material remains and cultural heritage. Field seasons from 2016 to 2019 recovered of several ceramic burial vessels from inhumation contexts, providing valuable …

Advancing Drug Discovery With Structural And Representation Learning Of Biological Systems, Mehdi Yazdani-Jahromi

Graduate Thesis and Dissertation post-2024

The rapid emergence of global health crises, exemplified by the Coronavirus Disease 2019 (COVID-19) pandemic, underscores the profound limitations of conventional drug discovery methods, which are inherently protracted and resource-intensive. A robust, accelerated response framework demands the sophisticated deployment of Artificial Intelligence (AI) and Machine Learning (ML) across the entire therapeutic pipeline. We propose a unified approach centered on \textbf{Structural and Representation Learning} to capture the intricate topological and chemical nuances of biological systems, thereby transforming complex molecular and genomic data into effective vector embeddings suitable for deep learning, enhancing predictive power, and crucially, interpretability. This dissertation details advancements across …

Addressing Vulnerabilities And Defesne Mechanisms In Transformer-Based Language Models, Mansour Al Ghanim

Graduate Thesis and Dissertation post-2024

Transformer-based Large Language Models (LLMs) have achieved remarkable success across Natural Language Processing (NLP) tasks, yet their increasing adoption has exposed critical vulnerabilities across input, model, and output layers. This dissertation systematically investigates these vulnerabilities and develops corresponding defenses, offering an end-to-end view of LLM security. First, we study model-level vulnerabilities through backdoor injection attacks that implant hidden malicious behaviors via hardware perturbations. We propose TrojBits, a hardware-aware attack that uses minimal bit-flips to insert trojans into models. Its three-stage design—Vulnerable Parameter Ranking, Hardware-Aware Attack Optimization, and Vulnerable Bit Pruning—success- fully compromises BERT and XLNet using only 64 parameters and …

Modeling A Moving Manipulator Platform In Ros 2 For Space Robotics, Bastian S. Weiss

Graduate Thesis and Dissertation post-2024

The control of mobile manipulators presents a unique challenge as the complexity of motion increases. ROS 2 provides the ability to create a digital twin of a physical system and control both simultaneously. This can be applied to mobile manipulators, allowing us to plan out complex movements and then simulate them virtually or physically. The Space ROS software is adapted for our Rapid Orbital Motion Emulator which is a 6 Degree of Freedom manipulator on top of a holonomic mobile base platform. Our custom Omni wheel controller running within the ROS 2 control package handles the kinematics for the base, …

Analysis Of An Aero-Engine Using Ammonia As A Fuel Carrier, Kangana Patel

Graduate Thesis and Dissertation post-2024

Numerous studies in the literature have explored the decarbonization of aviation. Some opt for sustainable aviation fuels (SAF), while others propose Hydrogen(H₂). Recent studies have also investigated Ammonia (NH₃) as a potential hydrogen fuel carrier. This study of NH₃ as a viable alternative fuel could help reduce overall emissions and overcome the challenges of H_2, which is highly flammable and difficult to store. This study aims to develop a core-engine model in a system simulation software called Simcenter Amesim. Results show a reasonable similarity between jet-A fuel and NH₃ cracked H₂ within …

Large Language Models And Networks: Edge Proposal, Synthetic Corpora, And Adaptive Node Generation, Nathan Gonzalez

Graduate Thesis and Dissertation post-2024

Natural Language Processing (NLP) and Artificial Intelligence (AI) have evolved from brittle rule-based engines through data-driven statistical models to today’s transformer architectures whose multi-head attention enables rich contextual reasoning. Against this backdrop, this dissertation unifies three complementary investigations that deepen our understanding of how large language models (LLMs) can reason over, generate, and strategically adapt language within networked settings. First, instruction-tuned LLMs are recast as latent-relationship detectors: by prompting models to hypothesize links between text-described entities, we recover edge sets that reconstruct social, thematic, and citation graphs with high precision, revealing how attention distributions encode topological cues. Recognizing the difficulty …

Performance Analysis Of A Turbofan Intergrated With An Entropy-Minimized Heat Exchanger For Ammonia-Powered Aviation, Lucas K. Cavalcante

Graduate Thesis and Dissertation post-2024

Ammonia has emerged as a promising aviation fuel due to its carbon-free emissions, well-established production methods, and transportation infrastructure. However, challenges with its slow chemical kinetics and low flame speed necessitate the partial cracking of Ammonia to introduce hydrogen as a combustion promoter. This study addresses these challenges by designing and optimizing a tube bank heat exchanger to transfer heat from the Cooled Cooling Air (CCA) of the High-Pressure Compressor (HPC) to the Ammonia stream, facilitating the necessary cracking. Preliminary heat transfer calculations indicate insufficient heat in the CCA to achieve cracking temperatures, prompting an alternative approach: splitting the Ammonia …

Investigating Algal Toxin Occurrence And Water Quality Correlations In A Freshwater–Estuarine System, Sunday Sam Egon

Graduate Thesis and Dissertation post-2024

Cyanobacterial toxins, particularly microcystin-LR (MC-LR), present persistent risks in freshwater and estuarine environments, especially under flow-regulated system where hydrodynamics and biology interact complexly. This study investigates the occurrence of MC-LR across a lake–estuary water matrixes, sampling 13 sites from Lake Okeechobee through Caloosahatchee canal, river, and estuarine zones from March to September. Water quality parameters including conductivity, temperature, dissolved oxygen, chlorophyll-a, and phycocyanin were measured in situ. The concentration of MC-LR, and nutrients (ammonium, orthophosphate, and nitrite concentrations) were measured in the laboratory. Multivariate correlation, principal component analysis, and multiple regression were employed to determine dominant drivers of toxin occurrence. …

Learning-Based Ethereum Phishing Detection: Evaluation, Robustness, And Improvement, Ahod Alghuried

Graduate Thesis and Dissertation post-2024

Phishing attacks continue to pose a significant threat to the Ethereum ecosystem, accounting for a major share of Ethereum-related cybercrimes. To enhance the detection of such fraudulent transactions, this dissertation develops a comprehensive framework for machine learning-based phishing detection in Ethereum transactions. The framework addresses critical aspects such as feature selection, class imbalance, model robustness, and the vulnerability of detection models to adversarial attacks. By systematically evaluating these key practices, this work contributes to the development of more effective detection methods. The first part of the dissertation assesses the current state of phishing detection methods, identifying gaps in feature selection, …

High-Efficiency Dual Active Bridge Converter With Stacked Secondary Phase For Next-Generation Ai Data Center Power Delivery, Anirudh Ashok Pise

Graduate Thesis and Dissertation post-2024

The rapid electrification of hyperscale AI data centers is accelerating a shift from legacy 48 V distribution ladders to ±400 V high-voltage DC backbones, concentrating residual loss and dynamic stress on a single isolated bidirectional interface. This work presents a high-density Dual Active Bridge with Stacked Secondary Phasing (DAB-SSP) architecture as a scalable GaN-based solution for 48 V–800 V rack-level power conversion. By stacking 400 V secondaries, the topology leverages 650 V GaN devices while satisfying 800 V system requirements, reducing clearance and creepage distances, minimizing magnetic path lengths, and embedding leakage inductance necessary for wide-range zero-voltage switching.

A rigorous …

Mems-Based Airflow Sensor For Apnea Detection Using Time-Of-Flight Technique, Parvin Akhkandi

Graduate Thesis and Dissertation post-2024

Sleep is essential for overall human health, influencing cognitive, emotional, and physical well-being. Respiratory-related sleep disorders, particularly obstructive sleep apnea (OSA), require early detection and effective treatment to prevent serious health complications. This dissertation explores innovative advancements in respiratory monitoring sensor development techniques, aiming to improve the diagnosis and management of sleep disorders. The sensor in this work incorporates a pair of Thin-film Piezoelectric on Substrate (TPoS) MEMS resonators operating at ~25 MHz, embedded in two oscillator circuits where frequency variations reflect changes in temperature and humidity caused by respiratory airflow. By precisely measuring airflow travel time between the resonators, …

Exploring The Genetic Structure And Evolutionary History Of Marine Fish Populations Across Micronesia, Pavithiran Amirthalingam

Graduate Thesis and Dissertation post-2024

Most marine species exist as metapopulations, composed of subpopulations inhabiting patchily distributed habitat. Connectivity among these subpopulations is primarily facilitated by larval dispersal via ocean currents. Understanding patterns of population connectivity is crucial for evaluating metapopulation dynamics and ensuring long-term population stability in managed species. Micronesia, a vast region in the North Pacific with hundreds of islands, is the northern extent for many tropical species and is thought to function as vital stepping-stones between the central and western Pacific. However, limited phylogeographic studies in the region have left significant knowledge gaps regarding metapopulation dynamics. Here I address this knowledge gap …

Advanced Large Language Model Approaches And Natural Language Processing Techniques To Improve Hate Speech Detection Using Ai, Saad Almohaimeed

Graduate Thesis and Dissertation post-2024

The proliferation of hate speech on social networks can create a significant negative social effect, making the development of AI-based classifiers that can identify and characterize different types of hateful speech in messages highly important for stakeholders. While this is a highly challenging problem, recent advances in language models promise to advance the state of the art such that even subtle and indirect forms of hate speech can be detected.

In this dissertation we present a series of contributions that improve different aspects of hate speech classification. We developed THOS, a hate speech dataset consisting of 8.3k tweets. Compared to …

Autoignition Delay Times For Reformate Gas Mixtures From Methane Gas Engines, Matthew Fraze

Graduate Thesis and Dissertation post-2024

Methane slip is a prominent issue in natural gas reciprocating engines that are used in transportation and marine applications. The incomplete combustion that results in methane slip can be resolved with the introduction of hydrogen within the combustion mixture to improve methane oxidation and further enable combustion within the engine crevices where methane has previously remained unreacted. Steam Methane Reforming (SMR) is a common method used to produce hydrogen and can be used to design an onboard device to reduce methane slip from reciprocating engines. The development of this reformer device requires the validation of high-fidelity chemical kinetic mechanisms at …

Integrated Performance-Based Framework For Evaluating Alternative Containment Systems In Subtitle D Landfills Managing Coal Combustion Products, Poyu Zhang

Graduate Thesis and Dissertation post-2024

Designing landfill containment systems that remain effective over decades to centuries poses significant challenges in geoenvironmental engineering, especially for landfills managing coal combustion products (CCPs) with complex hydraulic and chemical behaviors. This dissertation aims to develop an integrated, performance-based framework for evaluating landfill barrier systems by combining numerical modeling, experimental leaching tests, and machine learning prediction. The first component assesses the equivalency of the Florida double liner system relative to the Subtitle D composite liner mandated by the U.S. Environmental Protection Agency (US EPA). Using field leakage observations and COMSOL Multiphysics-based numerical simulations, results demonstrate that the Florida system, when …

Towards Left-Shifting Optimizations In The Hardware Design Methodology, Sanjay Gandham

Graduate Thesis and Dissertation post-2024

The exponential growth in computational demands has outpaced traditional Moore's Law scaling, necessitating innovations in hardware design methodologies. Traditional hardware design flows defer critical power, performance, and area (PPA) optimizations to late stages such as Logic Synthesis and Physical Design, creating prolonged feedback cycles that limit performance scaling. As designs grow in complexity, these late-stage optimization bottlenecks worsen, making it increasingly difficult to meet the desired PPA constraints for modern SoC designs with billions of transistors.

To address this limitation, this dissertation introduces the concept of left-shifting the hardware design process by enabling PPA optimizations at the earlier RTL stage. …

Exploring The Security Landscape Of Ar/Vr Applications: A Multi-Dimensional Perspective, Abdulaziz Alghamdi

Graduate Thesis and Dissertation post-2024

The rapid evolution of Augmented Reality (AR) and Virtual Reality (VR) technologies on mobile platforms has significantly impacted the digital landscape, raising concerns about security and privacy. As these technologies integrate into everyday life, understanding their security infrastructure and privacy policies is crucial to protect user data.

To address this, our first study analyzes AR/VR applications from a security and performance perspective. Recognizing the lack of benchmark datasets for security research, we compiled a dataset of 408 AR/VR applications from the Google Play Store. The dataset includes control flow graphs, strings, functions, permissions, API calls, hexdump, and metadata, providing a …

Characterization Of Turbulent Flames In Confined Combustors, Max Fortin

Graduate Thesis and Dissertation post-2024

As industry transitions to a net-zero carbon future, turbulent premixed combustion will remain an integral process for power generating gas turbines and are also desired for aviation engines due to their ability to minimize pollutant emissions. However, accurately predicting the behavior of a turbulent reacting flow field remains a challenge. To better understand the dynamics of premixed reacting flows, this study experimentally investigates the evolution of turbulence in a high-speed bluff-body combustor. The combustor operates across a range of equivalence ratios from 0.7-1 to quantify the role of heat release and flame scales on the evolution of turbulence as the …

Towards An Intelligent Speculative Software-Defined Networking, Ahmad Hariri

Graduate Thesis and Dissertation post-2024

Software-Defined Networking (SDN) separates the control and data planes, enabling better programmability of the control plane to predict, route, and schedule traffic at the data plane. Reactive SDN dynamically installs flow rules when a new flow arrives, making it adaptable to application dynamics. This design is well-suited for emerging low-latency applications like online gaming and AR/VR, which require millisecond-level response times for an acceptable quality of experience. However, Reactive SDN faces limitations, as each new flow triggers a miss, sending a Packet-in message from the switch to the SDN controller, increasing latency. To reduce delay, it's crucial to predict flow …

Characterizing Muscle Control As An Emg Visual Matrix, Peter W. Makary

Graduate Thesis and Dissertation post-2024

Evaluating electrical muscle activity patterns using surface electromyography (EMG) is a widely used technique to assess movement in research settings, but its use has not been translated to clinical settings. This study aims to investigate the potential of evaluating muscle activity control through a visual matrix display. EMG data from older and younger adults collected in a previous seated locomotor perturbation study was used to generate visual matrices representing their lower-limb ankle muscle activity. Visual matrices from each condition consist of 20x20 grids, where each axis represents activity from an individual muscle of muscle pairs, such as the tibialis anterior …

Investigating The Efficacy Of A Novel Law Enforcement Use Of Force Training Simulation, Justin D. Lane

Graduate Thesis and Dissertation post-2024

Realistic scenario based training (SBT) is an effective method used to prepare police officers for real world use of force incidents. Many law enforcement agencies are unable to host regular SBT due to time, cost, and facility constraints. A novel simulation training approach was investigated to determine if it may be a suitable alternative or supplement to traditional SBT at reduced cost, time, and increased frequency. This novel scenario based simulation training (SBST) environment combined physical and virtual elements into a single training scenario. A physical space, designed to depict the scene of a critical incident, was equipped with immersive …

Aluminum Alloys For Laser Powder Bed Fusion Additive Manufacturing And Process Optimization By Machine Learning, Kevin Graydon

Graduate Thesis and Dissertation post-2024

Laser powder bed fusion (LPBF) additive manufacturing is a promising manufacturing technology enabling enhanced design freedom through layer-by-layer production. However, traditional, wrought aluminum (Al) alloys suffer from solidification cracking during LPBF processing and the initial printer parameter optimization process required for every novel or untested material is time and resource intensive. There exists a need to formulate Al alloys for LPBF and a method to reduce the resources consumed during initial printer parameter optimization studies. Al-9.5Ce-xMo (x = 0.2, 0.6, 1.0 wt. %) alloys have been designed specifically for LPBF utilizing the CALPHAD approach. Additionally, a neural network model was …

Deepfake Detection And Defense: An Analysis Of Techniques And Robustness, Nigel Francis

Graduate Thesis and Dissertation post-2024

The rise of deepfake media powered by the abrupt and rapid advances in generative AI, has introduced serious concerns and challenges in digital content verification. Although there is a wide range of detection strategies currently in use and being proposed, many remain limited in their ability to generalize across manipulation types, defend against adversarial attacks, or operate efficiently in real world scenarios. This paper presents a survey of deepfake detection and defense techniques, organized into ten technical categories: frequency-based, spatial-based, temporal-based, physiological-based, multimodal detection, adversarial attacks and defenses, forensic noise-based methods, watermarking and prevention strategies and dataset contributions. Drawing on …

Increasing Bystander And Environmental Awareness Through Haptic Feedback, Jose-Valentin T. Sera-Josef

Graduate Thesis and Dissertation post-2024

The Chariot System is a bystander and environmental awareness system that uses vibrotactile haptic feedback delivered via a wearable. It was designed with the goal of using a subject's haptic channel to keep them aware of their surroundings thus freeing their visual channel to focus on and enjoy a virtual reality (VR) experience. This work contributes to the space of VR research that aims to keep users aware of the outside world while in VR, an area that typically focuses on using visual notifications but rarely applies haptics. A user study involving 24 participants was conducted comparing the Chariot System …

Modeling Data Metrics And Distributions For Representation And Efficient Transfer Learning, Li Ren

Graduate Thesis and Dissertation post-2024

Representation Learning and Transfer Learning are fundamental challenges in deep learning, particularly for downstream computer vision tasks. While existing research predominantly focuses on loss functions and network architectures, this dissertation explores how modeling distance metrics and data distributions can enhance representation and transfer learning efficiency in computer vision applications, specifically focusing on image captioning and retrieval tasks.

We make contributions primarily in three subareas. First, we develop an image captioning system that leverages geometric relationships among objects using a novel object graph structure, enhancing global visual representations for more accurate caption generation. We further incorporate adversarial learning to model data …

Down By The Tracks, Ova ‘Cross The Trussle: Exploring Systemic And Structural Barriers To Perinatal Healthcare Access For Black Women In Rural Alabama, Trinity C. Johnson

Graduate Thesis and Dissertation post-2024

This dissertation uses semi-structured interviews (n=42), participant observation, archival research, and storytelling, it employs frameworks of structural violence, intersectionality, and reproductive justice to situate Black women’s maternal health within broader histories of racialized medical neglect and reproductive governance. The study explores two questions: (1) What obstacles do Black women in rural Alabama face in accessing perinatal healthcare? (2) How do they navigate these barriers, and what strategies do they employ? A mixed-methods approach—including ethnography, quantitative analysis, and GIS—guides this research, with storytelling serving as both a methodological intervention and a challenge to dominant anthropological narratives. Findings highlight three major themes: …

Long Term Trends In The Reproductive Output And Phenology Of Three Marine Turtle Species, Jeena Prasertlum

Graduate Thesis and Dissertation post-2024

The UCF Marine Turtle Research Group (MTRG) has monitored sea turtle nesting along the Brevard County, Florida coastline for over 40 years, yielding one of the world's longest continuous sea turtle reproductive datasets. These data offer key demographic insights for conservation and species recovery assessments. Given climate change’s worsening effects, it is critical to analyze long-term trends in hatchling emergence success and phenology, which impact reproductive output, population growth, and hatchling dispersal. Here I compare the long-term reproductive output trends of loggerhead (Caretta caretta), green (Chelonia mydas), and leatherback (Dermochelys coriacea) sea turtles from …

Haptics Based Grasping Of Novel Objects Using Robotic Manipulator, Nicolas C. Leocadio

Graduate Thesis and Dissertation post-2024

This work is focused on extending and demonstrating the generalizability of previous interactive perception-based work done on a model-free, non-neural algorithm for robotic gripper envelopment of objects using haptic feedback. Using a 6-axis force and torque sensor placed behind the wrist of a robotic arm's manipulator, appropriate motions are generated in order to achieve contact with, align the manipulator with, and place the manipulator around some desired object in a position where grasp is possible by simply closing the manipulator. The algorithm relies on generating restoring (reactive) and frictional forces by rapid multidirectional oscillation of gripper fingers about an initial …

Enhancing The Hardness Of The Multispectral Zns Transparent Ceramics Through Ga2s3 Incorporation: A Comparative Study Of Vapor Transport Deposition And Solid-Solid Reaction Methods, Hayat Soufiani

Graduate Thesis and Dissertation post-2024

Multispectral zinc sulfide (MS-ZnS) is an important optical ceramic widely used in broadband imaging systems due to its wide transmission range and relatively low cost. However, its modest mechanical properties limit its durability in harsh environments. This dissertation investigates two solute strengthening methods, vapor transport deposition (VTD) and diffusion coupling (DC), to enhance the hardness of MS-ZnS windows through Ga2S3 incorporation while maintaining optical transparency. The VTD process successfully incorporated gallium in the subsurface, resulting in up to 120 % increase in microhardness compared to untreated samples, but with slight discrepancies in lateral gallium distribution. The DC technique demonstrated superior …

‘Who’S Your Daddy?’: Ddradseq Parentage In Broodstock-Offspring Pairs Of Diadema Antillarum In A Population Restoration Initiative, Bianca R. Rosendahl

Undergraduate Scholarship and Creative Works

In the early 1980s Diadema antillarum, or the long-spined urchin, was hit by a catastrophic mass mortality event across the western Atlantic and Gulf of Mexico, inclusive of the Florida Keys. The cause of this die-off event was likely triggered by a pathogen or environmental stressor that decimated native populations by up to 98%. The species has not yet fully recovered and is struggling with recruitment success which is why it may benefit from an ex-situ recovery program. In this study, I used single nucleotide polymorphisms (SNPs) to identify genetic variation and analyze parentage in a captive population of …