Digital Commons Network^™

Exploring Factors That Influence Artificial Intelligence Adoption In Banks And Credit Unions, Vijaya S. Tumma

USF Tampa Graduate Theses and Dissertations

The importance of Artificial Intelligence (AI) is exploding in the banking sector, fueled by enhanced productivity, improved efficiencies, and personalized services to the consumers. For credit unions, the adoption of AI technologies presents opportunities and challenges. This research explores the factors influencing AI adoption in the banking sector through the lens of Unified Theory of Acceptance and Use of Technology (UTAUT) framework. This study aims to explore the influence of key aspects of UTAUT model, Performance Expectancy (PE), Effort Expectancy (EE), Social Influence (SI), and Facilitating Conditions (FC) on the intention of AI adoption among credit unions, banks, and their …

Understanding And Alleviating The Impact Of Convective Weather To Airfield Efficiency, Huang Feng

USF Tampa Graduate Theses and Dissertations

Extensive research in predicting annual passenger throughput at airports has been conducted, aiming at providing decision support for airport construction, aircraft procurement, resource management, flight scheduling, etc. However, few studies have been conducted on how airport operational throughput is affected by convective weather in the vicinity of the airport and how to predict short-term airport operational throughput. Convective weather near the airport could make arrivals miss their positions in the arrival stream and reduce airfield efficiency in terms of utilizing runway capacities. This research leverages the learning-based method (MB-ResNet model) to predict airport hourly throughput. It uses the Hartsfield–Jackson Atlanta …

Spatiotemporal Dynamics Of Coastal Nekton Metacommunities In The Western Atlantic, Jonathan A. Peake

USF Tampa Graduate Theses and Dissertations

The processes of diversification, selection, dispersal, and stochastic drift can combine in many ways to assemble ecological communities. A key focus in ecology is to describe and quantify the relative influences of these processes for different systems and explain the mechanisms by which these processes produce patterns of community structure. To accomplish this goal, ecologists have developed several frameworks that seek to integrate processes that occur locally andregionally. The metacommunity concept treats local communities as pieces of a broader landscape that are connected through dispersal of their constituent organisms. Empirical studies have generally lagged behind the development of metacommunity theory. …

Balancing Context And Clarity Through Visualizations For Better Decision-Making, Bhavana Doppalapudi

USF Tampa Graduate Theses and Dissertations

In this era of a data driven world, the effective communication of data through visualizations is pivotal for converting complex information into accurate insights. Effective visualizations not only enhance users' comprehension of complex data but also induce trust, leading to better decision-making. This dissertation explores methods to improve trust in visualizations by providing additional context and enhancing their clarity, ensuring appropriate data interpretations and better decisions from users.

The work in the dissertation begins by examining the impact of scatterplots combined with statistical metrics such as accuracy on users' trust and decision-making focused on recommender systems. In addition, we show …

Coding Connections: Exploring Relationships Between Computer Science Learning And Mathematics Achievement In Secondary Education, Bradley Hayes

USF Tampa Graduate Theses and Dissertations

This dissertation in practice explores the intersection of computer science education, specifically computational thinking and programming, with mathematics achievement among 15-year-old students in selected English-speaking countries. The research addresses a gap in understanding whether skills developed through computer science can positively influence mathematics performance by assessing the extent to which learning in computer science transfers to mathematics.

To achieve this, a quantitative methodology was employed, incorporating pilot study data from a single school and large-scale survey data from the Programme for International Student Assessment (PISA) 2022. The analysis assessed the correlation between regular participation in programming activities and mathematics attainment, …

Analyzing And Extending Machine Learning Frameworks On High Risk Domains, Chengbin Hu

USF Tampa Graduate Theses and Dissertations

Machine learning (ML) has become a transformative force in high-risk domains such as genomics and cybersecurity, where accurate predictions and robust defenses are essential. This dissertation advances ML frameworks in these areas by developing methods to enhance predictive power in health applications and assess vulnerabilities in machine learning systems.

In the genomics field, the work addresses challenges in Non-Invasive Prenatal Testing (NIPT) of monogenic disorders by proposing a deep learning model that reconstructs the fetal genome using maternal plasma cell-free DNA (cfDNA) and parental whole-genome sequencing (WGS) data. This model achieves high accuracy in single nucleotide variation (SNV) prediction, surpassing …

Detecting Changes In Cancer-Related Cognitive Functioning Using Mixed-Effects Modeling And Study Design Simulations, Jonathan D. Clapp

USF Tampa Graduate Theses and Dissertations

The purpose of this dissertation is to utilize and compare two existing approaches for modeling nonlinear change in cognitive functioning in cancer patients over time and to assess alternative study designs to better capture chemotherapy-related cognitive impairment using simulation data. Using data from the Thinking and Living with Cancer study, trend analysis with orthogonal polynomials (OP) and piecewise (PW) regressions approaches used to analyze cognitive functioning trajectories. Simulated datasets tested the impact of study design parameters including measurement test-retest reliability (.6 or .8), sample size (100, 200, or 300), and the number of data collection time points (4 or 5) …

Control Of Multiagent Networks With Misbehaving Nodes, Emre Yildirim

USF Tampa Graduate Theses and Dissertations

The overarching objective of this dissertation is to introduce system theoretical frameworks in control of misbehaving multiagent networks subject to limited resources (i.e., limited number of driver nodes), and the presence of disturbances and/or uncertainties (i.e., misbehaving nodes). Below, we briefly summarize each problem studied in this dissertation with their proposed solutions.

Specifically, we first consider multiagent networks subject to misbehaving nodes over fixed undirected graph topologies. To address this problem, we apply proportional-integral control signals to a subset of nodes (i.e., driver nodes) in the network to address the undesired effects arising from misbehaving nodes. Since multiagent networks with …

Examining Emergency Remote Teaching And Undergraduate Academic Outcomes During The Pandemic, Melissa S. Ramge

USF Tampa Graduate Theses and Dissertations

The COVID-19 pandemic represented an extraordinary public health emergency, leading to the most profound disruption to education systems in modern history. As higher education institutions expand their post-pandemic online course catalogs, it is necessary first to explore the student outcomes and student perceptions of learning related to the pandemic. This study used an explanatory sequential mixed methods research design to assess the relationship between the COVID-19 pandemic and students’ perceptions of online learning. The purpose of the study was to examine what, if any, trends in undergraduate student course grades and student perceptions of the instruction occurred coincident with the …

Telu Activation Function For Fast And Stable Deep Learning, Alfredo Fernandez

USF Tampa Graduate Theses and Dissertations

We propose the Hyperbolic Tangent Exponential Linear Unit (TeLU), a neural network hidden activation function defined as $TeLU(x)=x \cdot tanh(e^x)$. TeLU’s design is grounded in the core principles of key activation functions, achieving strong convergence by closely approximating the identity function in its active region while effectively mitigating the vanishing gradient problem in its saturating region. Its simple formulation enhances computational efficiency, leading to improvements in scalability and convergence speed. Unlike many modern activation functions, TeLU seamlessly combines the simplicity and effectiveness of ReLU with the smoothness and analytic properties essential for learning stability in deep neural networks. TeLU’s ability …

Exploring Llm Integration And Its Influence On Agent Behavior And Productivity In Insurance Call Centers, Gerardo L. Wibmer Gonzalez

USF Tampa Graduate Theses and Dissertations

This study investigates the theoretical impacts of supportive AI tools, specifically Large Language Models (LLMs), on agent behavior and communication dynamics in call centers. While technological advancements have streamlined operations, limited research addresses the indirect ways these tools influence agent behavior. Using frameworks like context switching—the cognitive shift required when external stimuli prompt attention shifts—and the Hawthorne effect, where perceived observation modifies behavior, we examine how LLMs shape communication patterns. In call centers, this context switching occurs indirectly as agents adapt to AI note-taking features, whereas in industries like Architecture, Engineering, and Construction (AEC), automation tools prompt more immediate changes …

Emergent Magnetism And Magneto-Transport Phenomena In Helimagnet-Based Heterostructures For Spin-Based Device Applications, Nivarthana Waruni Yapa Abeyrathna Yapa Mudiyanselage

USF Tampa Graduate Theses and Dissertations

The study of magnetoresistance (MR) phenomena in magnetic materials has been pivotal for advancing magnetic sensors and spintronic devices. Helimagnets, characterized by a non-collinear arrangement of magnetic moments that form a helical or spiral configuration, present an intriguing avenue for MR-based device applications. Theoretical predictions suggest that the MR magnitude in the helimagnetic (HM) regime can exceed that in the ferromagnetic (FM) regime by over an order of magnitude. However, in metallic helimagnets like manganese phosphide (MnP), the MR in the HM phase remains modest (< 10%), limiting its applicability in MR devices. This dissertation investigates the emergent magnetism and magneto-transport phenomena in nanostructured MnP helimagnets and their interfaces with ferromagnets (Fe) and ferrimagnets (Fe3O4), aiming to enhance device performance.

In the first part of the dissertation, we present a novel approach to achieve …

Sea Surface Temperatures, The Gulf Of Mexico Loop Current, Hurricane Ian, And Their Junction, Alexander K. Nickerson

USF Tampa Graduate Theses and Dissertations

In the first chapter, a climatology was created for the deep Gulf of Mexico, West Florida Shelf, and Tampa Bay. These climatologies were compared to each other and to global temperatures. Seasonal variability is shown to be a product of bathymetry as well as latitude. In Chapter Two, using the self-organizing map neural network, patterns of the Gulf of Mexico Loop Current were analyzed to establish phases with necessary and sufficient conditions for translation from one phase to the next. Two separate patterns of evolution were established from these analyses, and empirical arguments are used to confirm the findings from …

Ai-Enabled Methodological Frameworks To Enhance The Robustness Of Machine Learning Models In Various Domains And Dynamic Environments, Mohammad Noroozi

USF Tampa Graduate Theses and Dissertations

This dissertation explores novel approaches to address complex challenges in various domains, focusing on ensuring that machine learning (ML) models can reliably detect both the data they are trained on and unseen data. Reliable detection of unfamiliar inputs is critical for deploying models in real-world applications, where performance can degrade when faced with novel or unexpected information. Each domain addressed in this work introduces a set of contributions aimed at enhancing model performance, broadening generalization through diverse training data, and improving robustness against adversarial attacks. Three sets of contributions are presented, each encapsulated in a chapter.

One of the contributions …

Satellite-Based Mapping Of Seagrass Habitats And Coastal Wetlands In The Us Gulf Of Mexico And Florida: A Remote Sensing Approach For Environmental Conservation And Management, Luis Lizcano-Sandoval

USF Tampa Graduate Theses and Dissertations

Coastal wetlands and seagrass meadows provide multiple services to nature and human society, but their spatial and temporal changes are often not well characterized. They are habitats that are threatened by natural and human causes such as changes in water quality, urban development, pollution, and climate change, among others. Coastal ecosystems rely on water quality and sea level variability for survival, and changes in those variables and their potential effects are of concern to many coastal communities. Satellite remote sensing is a powerful tool that has been used for mapping ecosystem distribution on land and in the ocean. Satellite sensors …

Characterizing The Temporal And Nearshore To Offshore Spatial Scale Of Phytoplankton Assemblages Within The Florida Keys National Marine Sanctuary, Sebastian P. Di Geronimo

USF Tampa Graduate Theses and Dissertations

This study investigates the spatial and temporal dynamics of phytoplankton communities within the Florida Keys National Marine Sanctuary (FKNMS) from 2016 to 2021. Phytoplankton biomass and community composition were analyzed across four seasons and from both Nearshore (within 2 km of shore) and Offshore (8 to 10 km from shore) locations along the Florida Keys from Miami to the Western Sambo Ecological Reserve. High-performance liquid chromatography (HPLC) quantified chlorophyll-a (tchla) as a proxy for bulk biomass, and other pigments. CHEMTAX, a chemotaxonomic identification method, was employed to estimate phytoplankton community composition. Our results revealed significant seasonal variability in bulk biomass, …

On The Role Of Prediction In Streaming Hierarchical Learning, Ramy Mounir

USF Tampa Graduate Theses and Dissertations

In today's world, AI systems need to make sense of large amounts of data as it unfolds in real-time, whether it's a video from surveillance and monitoring cameras, streams of egocentric footage, or sequences in other domains such as text or audio. The ability to break these continuous data streams into meaningful events, discover nested structures, and predict what might happen next at different levels of abstraction is crucial for applications ranging from passive surveillance systems to sensory-motor autonomous learning. However, most existing models rely heavily on large, annotated datasets with fixed data distributions and offline epoch-based training, which makes …

Zooplankton Abundance, Composition, And Diversity And Associated Environmental Factors Across The Gulf Of Mexico And Along The Northwestern Coast Of Cuba, Shaniqua J. Gladney

USF Tampa Graduate Theses and Dissertations

My research focused on characterizing the zooplankton community of two regions in the Gulf of Mexico (GOM) along two transects, one across the middle of the GOM from Tuxpan, Mexico, to St. Petersburg, FL, and one along the northwestern coast of Cuba. Zooplankton bongo samples and associated environmental and physical processes were examined. These data were collected during two Center for Integrated Modeling and Analysis of Gulf Ecosystems (C-IMAGE) expeditions aboard the R/V Weatherbird II, during September 2015 (GOM) and May 2017 (Cuba). A Zooscan imaging system was used to identify and enumerate zooplankton taxa. Then zooplankton abundance, composition, and …

Enhancing Post Silicon Visibility Using Language Modelling Techniques, Nathaniel Joseph Fender

USF Tampa Graduate Theses and Dissertations

The debugging phase is a critical time in the development of a new system on chip product. Specifically, the post-silicon validation phase is one of the most important, as it allows engineers to test the behavior of a device in a real world setting. However, the issue of noisy or incomplete data is a frequent issue when attempting to debug an SoC design during this step. This thesis examines the utility of utilizing machine learning models for the purpose of repairing missing data in a system trace. We trained various models using the transformer architecture to identify missing data in …

Predicting Human Chess Moves With Large Language Models, Benjamin Kreiger

USF Tampa Graduate Theses and Dissertations

As artificial intelligence (AI) surpasses human performance in more tasks, the interest in leveraging and collaborating with this technology for greater productivity continues to grow. However, the black-box nature of current AI can make it difficult to interpret and unsuitable to perform tasks that are more complex and require human intuition. This has led to the pursuit of AI systems that can model individual behavior. Chess offers an ideal environment to explore this task due to its complexity, structure, and the abundance of data containing unique human decision-making examples. Given that a chess game can be fully represented with text, …

Optimizing Hidden Neurons In Multilayer Perceptrons With A Hybrid Feature Selection Approach For Classification Tasks, Susmitha Boyidapu

USF Tampa Graduate Theses and Dissertations

Identifying the ideal number of hidden neurons in multilayer perceptron (MLP) networks remains an ongoing challenge in the field of machine learning. Our research presents an innovative strategy to address this issue by combining data preprocessing, feature selection by a hybrid method, and a methodical approach to MLP architecture design. The extracted features through feature selection help to construct MLP architecture and implement k-fold cross-validation along with early stopping mechanism for finding optimal number of hidden neurons. Our method's key innovation is its capacity to dynamically adjust the hidden neuron count while optimizing model performance. Through iterative refinement of the …

Side-Channel Analysis Platform For A Hardware Implementation Of Fips 203 (Crystals-Kyber), Mohamed Mossad

USF Tampa Graduate Theses and Dissertations

In 2024, NIST selected the Post-Quantum Cryptography algorithm CRYSTALS-Kyber for standardization as a public-key encryption, key establishment scheme. CRYSTALS-Kyber was standardized under the Federal Information Processing Standard (FIPS), specifically FIPS 203. FIPS standards represent a set of guidelines, developed by NIST, for secure data handling in federal information systems, mandating cryptographic algorithms that protect sensitive information. This highlights the importance of identifying potential vulnerabilities in the algorithm and assessing how CRYSTALS-Kyber implementations react to hardware side channel attacks. Previous research identified several vulnerabilities in implementations of CRYSTALS-Kyber in software, which were addressed in subsequent releases. This thesis focuses on expanding …

Biodegradable And Green Controlled Release Formulations For Fertilizers, Srivalli Sriperambudur

USF Tampa Graduate Theses and Dissertations

This study aims to address environmental concerns associated with conventional fertilizers, and the persistence of synthetic polymers in soils and investigates the use of biodegradable cellulose esters as sustainable coatings for controlled-release fertilizers (CRFs). The focus is mainly on the synthesis of biodegradable cellulose esters as sustainable coatings for CRFs and applied as a coating on urea fertilizer prills to control nutrient release. The synthesized biodegradable coatings aim to provide effective nutrient delivery over time while minimizing environmental impact. Cellulose esters were synthesized using both saturated and unsaturated fatty acids, followed by crosslinking using thiols. Characterization techniques like FTIR, DSC, …

Learning Peer Support Interactions Via Bi-Lstm Graph Neural Networks For Suicide Risk Prediction, Harikrishna Marampelly

USF Tampa Graduate Theses and Dissertations

Suicide prevention through early detection using social media data has been widely studied. However, the critical role of peer support interactions among individuals with similar mental disorders has not been deeply investigated or explored. In this study, we explore peer interactions in online communities for individuals with bipolar disorder and leverage this information to predict suicide risk levels. We propose a model that uses contextualized posts and comments along with their sentiment features. By embedding these features into a peer support network, our model captures peer interactions and predicts suicide risk levels using the bidirectional LSTM Graph Neural Networks (Bi-LSTM …

Vulnerability Discovery On Broadcom Devices With Qemu, Christian Shane Douglas Little

USF Tampa Graduate Theses and Dissertations

System-on-chip (SoC) devices in the form of routers, access points, and smart appliances have become an increasingly prevalent aspect of our infrastructure. Performing vulnerability research on these devices may be exceptionally difficult given the on-hardware limitations, inability to guarantee certain aspects of the system, and lack of information when the appliance’s source code is not available.In this thesis, we present Broadfeather, a QEMU-based emulation target, and framework for dynamic analysis and instrumentation. While Broadfeather supports data types shared across many chips under the Broadcom umbrella, it is primarily geared towards the BCM6755 Wi-Fi 6E capable chipset. Broadfeather also features the …

Revealing Grounding Zone And Subglacial Processes At Cosgrove Ice Shelf: Seafloor Analogs For The Modern Antarctic Ice Sheet, Rosemary E. Burkhalter-Castro

USF Tampa Graduate Theses and Dissertations

The future loss of ice in West Antarctica is one of the biggest uncertainties in projections of global sea level rise. The West Antarctic Ice Sheet (WAIS), home to Cosgrove Ice Shelf (CIS), contains enough ice to raise global sea level by more than three meters if it collapsed completely. Accurately predicting how much sea level change may occur from a future collapse of WAIS is of particular interest to low-lying, heavily populated coastal areas such as Florida. For this reason, it is important to understand the processes underpinning rapid changes and retreat at marine-terminating glaciers in the Amundsen Sea. …

Assessing The Impact Of Ai Assisted Software Development And User Experience Of A College Football Simulation Game: A Study Of Player And Industry Professional Perspectives, Augustus J. Scarlato Iii

USF Tampa Graduate Theses and Dissertations

This research examines the use of Artificial Intelligence (AI) in the design of video games, specifically the development of a college football simulation game. This study documents the creation of an alpha version college simulation game assisted by Open AI’s Chat GPT 4.0 API, to potentially improve game development, gameplay realism, and user interaction. The study then assesses how both student players and industry professionals perceive AI-enhanced gaming, emphasizing the usability, gameplay experience, and overall quality of the game using a Likert scale survey. The analysis also highlights differences in perceptions between students and industry professionals, with the latter group …

Performance Comparison Of Two Binary Diagnostic Tests With Incomplete Data, Meng Wang

USF Tampa Graduate Theses and Dissertations

Diagnostic tests are medical examinations essential for patient care, as they help detect the presence of diseases and rule out certain conditions. However, diagnostic tests can make errors, and misdiagnoses resulting from these errors can potentially lead to treatment delays or inappropriate treatments. Therefore, comprehensively evaluating the performance of the diagnostic tests and ensuring test accuracy is a fundamental priority in diagnostic studies. When a new diagnostic test is developed, it is customary to compare it with an existing test in a paired study. However, incomplete test results commonly occur in practice. The maximum likelihood-based method can be used to …

The Attitudes, Perceptions, And Motivations Of Student-Athletes In Intermediate Algebra, Alexandra Green

USF Tampa Graduate Theses and Dissertations

This research delves into the attitudes, perceptions, and motivations of student-athletes enrolled in Intermediate Algebra courses. It focuses on the distinct challenges they encounter in juggling academic and athletic responsibilities at a southeastern university in the United States. Employing a mixed-methods approach, the study gathered quantitative data through surveys to assess students' attitudes, perceptions, and motivations toward learning mathematics, as well as to determine the influence of their instructors on these factors. The qualitative segment provided deeper insights into their personal experiences, shedding light on their prior encounters with mathematics, their most memorable moments during the current course, and their …

Essays On Information Technology In Healthcare, Gleb Zavadskiy

USF Tampa Graduate Theses and Dissertations

Information technologies (IT) and information systems (IS) have profound significance across various sectors of society, including healthcare, business, education, government, and beyond. First, IT facilitates instant communication globally through email, messaging apps, video conferencing, and social media, revolutionizing how individuals and organizations interact, collaborate, and share information (Hacker et al. 2020; Tang and Hew 2020).

Secondly, the Internet and digital libraries provide worldwide access to vast amounts of information, what makes knowledge and education available to everyone, empowering individuals to learn and stay informed on diverse topics (Haleem et al. 2022). Another aspect of IT systems in various industries is …