Digital Commons Network^™

Advances In Cluster Detection Via Order Statistics, Renewal Processes And Model Selection Methods With Applications To Financial Forensic Statistics, Kwabena Ansong Boateng

Theses and Dissertations

This dissertation develops a novel method to detect clusters that appear in financial transaction data that are linked to fraud. The methods utilize properties of order statistics, renewal processes, and model segmentation. We apply the Minimum Description Length (MDL) principle to select the best model. Our technique makes use of a genetic algorithm to determine the number and location of anomalous cluster(s) embedded in a set of financial transactions. We apply our novel cluster detection method to the Paycheck Protection Program (PPP), a U.S. Government initiative launched in April 2020 amid the COVID-19 pandemic to aid small businesses affected by …

The Effect Of Teacher Efficacy On Career Technical Education Program Participation In Select Mississippi Schools, Lori Parkman Nail

Theses and Dissertations

The primary purpose of this study was to determine to what degree school staff understood the benefits and outcomes of Career Technical Education programs and how this level of understanding impacted Career Technical Education program enrollment in five selected schools in Mississippi. The levels of understanding keyed in on three pivotal factors. First, perceptions of the availability and accessibility of Career Technical Education programs to students in the participating schools were measured. Second, the perceptions of the benefits students reap from being involved in Career Technical Education programs were examined. Next, the perceptions of the benefits Career Technical Education programs …

Determining The Impact Of Incorporating National Board Standards During Pre-Service Experiences On Teacher Success, Future National Board Certification, Student Growth, And Teacher Retention., Michelle Leigh Stubbs

Theses and Dissertations

This project addresses the critical issue of teacher attrition and retention by incorporating National Board standards into pre-service teacher preparation programs. Nationwide, school districts face significant shortages of qualified teachers, particularly in mathematics, science, and special education. Research reveals that the Southern region of the United States experiences the highest attrition rates (Ingersoll et al., 2018). An analysis of data indicates that the key issues contributing to teacher attrition and retention include stress from high-stakes testing, salary deficits, and inconsistent support. States have the autonomy to address teacher retention issues and allocate federal funds for these initiatives such as high-quality …

Physics-Based And Data-Driven Constitutive Modeling Of Unsaturated Soils Under Multi-Physics Loading Conditions, Mohsen Ajdari

Theses and Dissertations

Understanding the behavior of unsaturated soils under multi-physics loading is crucial for addressing several challenges related to emerging geo-energy technologies, climate change, and geohazard mitigation. The main objective of this research is to enhance the state of the art for modeling unsaturated soils under multi-physics loading conditions through physics-based and data-driven approaches. Toward this objective, this dissertation is divided into two distinct yet interrelated parts. In the first part (Chapters 2, 3, and 4), thermodynamic principles are employed to develop new constitutive functions to describe soil suction and stress-dilatancy of unsaturated soils under temperature-dependent conditions. Building upon the insights gained …

Multi-Cloud Identity Security Utilizing Self-Sovereign Identity, Morgan Lee Reece

Theses and Dissertations

With the increasing use of multi-cloud environments, security professionals face challenges in configuration, management, and integration due to uneven security capabilities and features among providers. As a result, a fragmented approach toward security has been observed, leading to new attack vectors and potential vulnerabilities. Other research has focused on single-cloud platforms or specific applications of multi-cloud environments. Therefore, there is a need for a holistic security and vulnerability assessment and defense strategy that applies to multi-cloud platforms. This dissertation explores risk and vulnerability analysis to identify attack vectors from software, hardware, and the network, as well as interoperability security issues …

Impact Of Cascading Optical Processes On Spectroscopic Quantification Of Absorption And Scattering Properties Of Samples, Pathum Dilshan De Silva Wathudura

Theses and Dissertations

Light absorption and scattering are fundamental properties of matter, particularly significant in nanoscale materials. This study explores how cascading optical processes affect the experimental measurement of absorption and scattering intensities, along with scattering depolarization, in samples containing absorbers and scatterers. The experimental UV–vis spectrum results from the combination of absorption and scattering extinction spectra. However, the maximum measurable extinction may be limited by forward scattered light. Light absorption decreases the intensity of scattered light and influences scattering depolarization. While scattering can reduce absorption along the optical path to the detector, sufficient absorption of the scattered light may counterbalance this effect, …

Analysis Of Cortical Evoked Auditory Response Detection In Adults Using Machine Learning, Pranavi Beerelli

Theses and Dissertations

This study focuses on the use of machine learning (ML) techniques to automate the detection of Cortical Evoked Auditory Responses (CEARs), which are key in understanding how the auditory cortex processes sound stimuli. Traditionally, analyzing these auditory responses has relied on manual interpretation by audiologists, a process that can introduce variability and human error, particularly in complex cases. To address this challenge, the research utilizes advanced deep learning models, including Convolutional Neural Networks (CNNs), Long Short Term Memory (LSTM) networks, and Bidirectional LSTM (BiLSTM) architectures, to analyze Electroencephalography (EEG) data and classify the presence or absence of auditory responses automatically. …

Serving Students With Specific Learning Disabilities: Identifying Prevalent Eligibility Practices In Mississippi, Phyllis Marie Yucatonis

Theses and Dissertations

This study investigated the determination eligibility practices for Specific Learning Disabilities (SLD) within Mississippi school districts, framed by the Individuals with Disabilities Education Act 2004 (IDEA) and Mississippi Department of Education (MDE) guidelines. A survey of 139 educational professionals in Mississippi offered insights into the methods preferred for determining SLD eligibility and the decision-making factors at the district level. The study examined commonly used methods for determining SLD, the challenges faced during the process, the need for professional development, the impact of educator demographics on SLD determination eligibility and general satisfaction with the SLD eligibility determination process. It also explored …

Evaluation Of The Physical And Mechanical Properties Of Hardwoods And Softwood Cross Laminated Timber, Richard Olakunle Omotayo

Theses and Dissertations

This thesis is focused on the evaluation of cross laminated timber made from southern pine lumber and low-grade hardwood species (which included red oak, sweetgum, and yellow poplar) growing in the southeastern part of the United States. The properties of interest were density, modulus of elasticity (MOE), modulus of rupture (MOR) and dynamic modulus of elasticity (dMOE). Mechanical testing using static bending technique was used to determine MOE and MOR and non-destructive technique (NDT) was used to predict dMOE. The results from mechanical testing showed that over 98% of low-grade hardwood CLT samples exceeded 8300 MPa. This proved the suitability …

Investigations Of Guayule Resin As A Wood Protectant, Queen Aguma

Theses and Dissertations

Wood products are subject to deterioration by numerous processes, as a biological material. Today, a significant area of preservative development is the utilization of existing plant residues that have water-repelling, antibacterial, antifungal, and insecticidal properties as components of wood protectants. Parthenium argentatum, known as guayule (wy-oo-lee), is a promising wood protectant due to its resin. However, the resin's complicated chemical, physical, and solvent-resistant properties make it difficult to use. The primary goal of this study was to eliminate the use of any solvent and rely on thermal modification only to fractionate usable compounds from guayule resin. Utilizing heated dialysis and …

Deep Learning - Based Automated Detection And Classification Of Foreign Materials In Poultry Using Color And Hyperspectral Imaging, Rohini Maram

Theses and Dissertations

This thesis explores the use of Deep learning for detection and classification of small foreign materials (FMs) in poultry meat using color and hyperspectral imagery (HSI). The study employs You only look once (YOLO) object detection models on color images for precise localization, and one-dimensional convolutional neural network (1D CNN), two-dimensional convolutional neural network (2D CNN) was used on HSI (600 – 1700 nm) for classification. Twelve different FMs commonly known as polymers including PVC, PET, LDPE and HDPE, were examined using 52 color and 52 hyperspectral images. Four YOLO models (v5x, v7x, v8x, v10x) were implemented, trained, tested and …

Survival Analysis In The Presence Of Independent Or Dependent Censoring, Sakie Jaladha Arachchige

Theses and Dissertations

This dissertation has three parts. The first part proposes a two-stage estimation procedure for a copula-based model with semi-competing risks data, where the nonterminal event is subject to dependent censoring by the terminal event. Under a copula-based model, the marginal survival functions of individual event times are specified by semiparametric transformation models, and a parametric copula function specifies the between-event dependence. The parameters associated with the marginal of the terminal event are first estimated, and the marginal parameters for the non-terminal event time and the copula parameter are second estimated via maximizing a pseudo-likelihood function based on the joint distribution …

On Topological Measures And Network Vulnerability Patterns: A Review And Comparative Analysis, Saviz Saei

Theses and Dissertations

Despite much hope for climate change to slow down or even reverse, younger generations face a future overshadowed by extreme events. The indisputable reality is that unless the United Nations establishes comprehensive and sustained climate justice policies, children today will experience five times more extreme events than those that took place a century ago. On Monday, July 3rd of 2023, an unprecedented peak in global temperatures was documented, marking the highest global temperature ever recorded, as the U.S. National Centers for Environmental Prediction reported. These increasing temperatures indicate the ongoing and intensifying phenomenon of climate change, which amplifies the frequency …

Improved Vector Pruning For Partially Observable Markov Decision Processes, Thomas Jonathan Bowman

Theses and Dissertations

Exact dynamic programming algorithms for planning problems that are represented as partially observable Markov decision processes rely on a subroutine that removes, or ``prunes", dominated vectors from sets of vectors that represent piecewise-linear and convex value functions. The classic vector pruning subroutine solves one linear program per vector, where the number of variables is equal to the size of the state space and the number of constraints is equal to the number of vectors shown so far to be undominated. Thus, its scalability is limited not only by the number of linear programs it solves, but especially by their size. …

Mapping Invasive Aquatic Plants Using Deep Learning Classification And Segmentation Models, Ruchitha Yadav Prakash

Theses and Dissertations

High-resolution imagery from Uncrewed Aerial Systems (UAS) equipped with multispec tral(MSI) and RGB sensors can be used for mapping invasive aquatic species. This research employs classification models like CNN, AlexNet, VGG16, ResNet, and DenseNet, alongside segmentation models such as FCN, U-Net, SegNet, DeepLabV3+, and Mask R-CNN, to analyze and visualize four invasive aquatic species including Cuban Bulrush (Oxycaryum cubense), Water Hyacinth (Eichhornia crassipes), Juncus (Juncus effusus), and Cattail (Typha latifolia). Among the tested models, ResNet (97.44% for RGB)andDenseNet(96.20%forMSI)exhibitedthehighest classification accuracies. For segmentation tasks, FCN (75.5% for RGB) and SegNet (79.6% for MSI) excelled. The study also examines the computational complexity …

Synthesis Of Glycan Antigens For Therapeutic Antibodies And Glycoconjugate Vaccines, Synthesis Of An Inflammatory Pentasaccharide, And Design And Synthesis Of Ceragenins For Use In Medical Devices And Osteomyelitis, Shawn Gubler

Theses and Dissertations

A promising approach in combatting infections from antibiotic resistant bacteria/fungi is using therapeutic monoclonal antibodies (mAbs) that can bind the outer surfaces of microbes. We have developed a synthetic vaccine platform that can rapidly generate mAbs against surface glycans from desired strains, such as lipopolysaccharides and capsular polysaccharides. To create therapeutic antibodies against Ruminococcus gnavus, Burkholderia multivorans, and Staphylococcus aureus, we synthesize glycan antigens from each strain that are necessary for use in our glycoconjugate vaccine platform. Blooms of the Gram-positive bacterium Ruminococcus gnavus have been correlated with inflammatory bowel disease, and recently a polysaccharide produced by this organism was …

Sulfhydryl-Containing Compounds And Iron Homeostasis: Mechanisms Of Labile Iron Pool Dysregulation And Consequences For Cellular Health, Steven Lowell Mann Christiansen

Theses and Dissertations

While iron's unique coordination chemistry and redox activity have resulted in its broad incorporation into biological systems, dysfunction in key iron-regulating mechanisms can transform these biochemical advantages into factors that contribute to several pathological conditions. This dissertation explores the complex interactions between iron and sulfhydryl-containing compounds (SCCs), with particular focus on glutathione's (GSH) role in stabilizing iron reactivity juxtaposed with homocysteine's (Hcy) ability to disrupt iron homeostasis and promote disease progression. Through rigorous biochemical analysis and experimentation, we demonstrate that Hcy disrupts iron regulation through several mechanisms: (1) Hcy competitively displaces protective iron-binding molecules in the Labile Iron Pool (LIP), …

The Learner's Experience Using Augmented Reality, Michael Lionel Atkinson

Theses and Dissertations

Augmented reality (AR) is a promising learning tool that has the potential to improve learning experiences in unique ways. While AR has been extensively studied in terms of learning outcomes, relatively little has been researched about the qualitative aspect of the learner experience and how that may inform better instruction. This multi-article, journal-ready dissertation addresses this gap. The first article is a review of the literature on what researchers have learned about AR in learning, systematically identifying six areas that appear salient yet underrepresented in the literature. The second article expands further upon this theme by conducting a qualitative study …

Enhancing Cybersecurity Strategies Through Automated Cti Extraction, Risk Prioritization, And Privacy-Conscious Information Sharing, Spencer Rian Massengale

Theses and Dissertations

Cybersecurity operations require the ability to collect and analyze large amounts of cyber threat intelligence (CTI) to assess risks and formulate defensive strategies against emerging threats. This task has become increasingly complex due to the rapid evolution of cyber threats and the growing volume of unstructured, natural-language CTI sources. The scale of data and analysis needed to utilize CTI effectively far exceeds humans' manual capacity, especially for organizations with limited resources. This research focuses on leveraging Large Language Models (LLMs) and machine learning techniques to enhance CTI extraction, risk assessment, and data sharing. We utilized LLMs to automate the extraction …

Raman Characterization Of Elastomeric Materials, Shane Michael Drake

Theses and Dissertations

Raman spectroscopy is often used for material identification but may be used for characterization. This document focuses on using Raman measurements to characterize polymers, specifically the curing and aging mechanism for hydroxyl-terminated polybutadiene (HTPB). HTPB is a well-known elastomer used in sealants, non-slip surfaces, rocket repellants and plastic explosives. Raman spectroscopy was used to monitor urethane linkages that form from curing HTPB with a cyanate curing agent to suggest a curing mechanism for HTPB. Since this polymer is highly susceptible to oxidation, the methods used to characterize curing were also used for the aging process to see if urethane breakages …

Physics-Informed Transfer Learning For Process Control, Samuel Arce Munoz

Theses and Dissertations

In the realm of process control, managing complex systems with limited prior knowledge presents significant challenges, particularly in environments where traditional mechanistic models are either unavailable or computationally prohibitive. This work explores the integration of deep transfer learning with system identification and Model Predictive Control (MPC) to develop control strategies that are both data-efficient and computationally streamlined. Initially, Long Short-Term Memory (LSTM) networks are employed to create approximate MPC controllers by leveraging transfer learning from a source system to a target system, demonstrating that transfer learning can achieve comparable performance to traditional MPC methods with reduced training data. Building upon …

Enhancement Of Ambient Air Quality Index Forecasting Using Optimized Ensemble Model, Vanitha M

Theses and Dissertations

Forecasting ambient air quality is essential for environmental sustainability and public health, especially in heavily populated regions such as China, India, and the United States where air pollution remains a serious concern. Traditional forecasting models often struggle to accurately represent air quality data because of its complex patterns and nonlinear interactions. To address these challenges and improve forecast performance, this research proposes a comprehensive strategy that integrates parallel heterogeneous ensemble modeling with Bayesian optimization.

The study begins with a seasonal machine learning–based imputation technique (SeasonalMLImpute) designed to handle missing data in meteorological and air quality parameters. This method is evaluated …

Sd-Weat: Towards Robustly Measuring Bias In Input Embeddings For Artificial Intelligence Language Models, Magnus Gray

Theses and Dissertations

Artificial intelligence (AI) is rapidly transforming industries and markets, from healthcare to entertainment, revolutionizing decision-making processes. However, as AI grow more influential, they also risk amplifying existing biases, potentially leading to harmful consequences. Recent advancements in large language models (LLMs), such as GPT-4 and Llama, have heightened concerns about bias in natural language processing (NLP) tasks, driving the need for robust methods to detect and mitigate bias. Current approaches, such as the Word Embedding Association Test (WEAT) and its sentence-level extension the Sentence Encoder Association Test (SEAT) often fall short in capturing the nuances of biases in the input embeddings …

Role Of Multimedia And Conative Words In Online Socio-Political Mobilization, Mainuddin Shaik

Theses and Dissertations

In the digital age, social media platforms have become vital for political mobilization, enabling rapid information dissemination, real-time engagement, and virtual community formation. This dissertation presents a novel multi-method framework to analyze mobilization dynamics on multimedia-centric platforms, focusing on Instagram, X (formally Twitter), and TikTok. The study examines Brazilian political protests (November 2022–February 2023) and disinformation and anti-disinformation campaigns during Taiwan's 2024 presidential elections. The research applies a multi-method framework to analyze mobilization characteristics in Brazilian and Taiwanese social movements, grounded in the Diffusion of Innovations (DOI) theory. This approach assesses how multimedia formats—images, sidecars, and videos—influence cognitive mobilization at …

The Raye Montague Digital Collection, Paola Cavallari

Theses and Dissertations

The Raye Montague Digital Collection consists of 735 items that have been digitized and catalogued to tell the story of Raye Montague, the United States Navy hidden figure. The Collection addresses the need to remedy decades of historical invisibility Raye Montague was subjected to because of gender and racial discrimination despite her fundamental contributions to U.S. history. This thesis details how the Collection inserts Raye Montague within a broader historical context that uncovers the role Black women had as federal employees in the U.S. military during and after World War II. It also explains how, in designing the Collection, I …

Lstm-Transformer Based Robust Hybrid Deep Learning Model For Financial Time Series Forecasting, Md R. Kabir

Theses and Dissertations

The inherent challenges of financial time series forecasting demand advanced modeling techniques for reliable predictions. Financial data, belonging to the category of multimedia data, contains an extensive quantity of information that is widely used for data analysis and decision-making tasks. Effective financial time series forecasting is crucial for financial risk management and the formulation of investment decisions. Accurate prediction of stock prices is a subject of study in the domains of investing and national policy. This problem appears to be challenging due to the presence of multi-noise, nonlinearity, volatility, and chaotic natures in stocks. This paper proposes a novel financial …

A Comparative Study Of Toxicity Propagation Using Epidemiological Models, Niloofar Yousefi

Theses and Dissertations

This dissertation explores the spread and intensity of toxic content on social media, with a particular focus on platform X (formerly known as Twitter). The growing presence of toxic interactions has sparked concerns about their effects on public discourse. Utilizing epidemiological models such as SIR (Susceptible, Infected, Recovered), SIS (Susceptible, Infected, Susceptible), SIRS (Susceptible, Infected, Recovered, Susceptible), SEIRS (Susceptible, Exposed, Infected, Recovered, Susceptible), and the novel SEImIhRS model (Susceptible, Exposed, Moderate Infected, High Infected, Recovered, Susceptible), which are traditionally used to study infectious disease dynamics, this dissertation examines toxicity in platform X discussions. A key contribution of this work is …

Nonlinear Thomson Scattering In An Intense Tightly Focused Ultrashort Laser Pulse, Yance Sun

Theses and Dissertations

We investigate nonlinear Thomson scattering generated by intense laser pulses focused in noble gases. Electrons are ionized from low-density helium or argon early during the laser pulse and scatter light out the side of the focus. We measure fundamental, second, and third harmonic light focus over nearly the entire emission sphere. We investigate the influence of electron bunching, when individual atoms ionize multiple electrons sequentially as the laser field increases early during the laser pulse. Simulations suggest that correlation between electron positions for the ten electrons ionized from argon should distort the spatial pattern of nonlinear Thomson scattering relative to …

Understanding The Role Of The Heat Transfer Coefficient Between Tool/Workpiece Interface During Friction Stir Welding, Matthew Goodson

Theses and Dissertations

The heat transfer coefficient is a key parameter in modeling friction stir welding and has yet to be measured experimentally. The importance of this parameter was shown through validating friction stir models on both sides of the tool/workpiece interface. Both a transient plunge and a steady state model were validated by matching experimental temperatures. The steady state tool temperatures were matched (with in 2.5%), but the steady state workpiece temperatures were off by around 20%. The transient model of the FSW plunge showed the effect of varying the ℎ𝑊/𝑇 on workpiece temperatures and tool temperatures. Two methods were looked at …

Investigating The Efficacy Of Prompt Engineering Techniques For Research Survey Paper Generation Using Large Language Models, Innocent Obed Awidi

Theses and Dissertations

The recent advancements in large language models (LLMs), such as ChatGPT, offer promising potential for automating academic writing tasks. This study investigates the efficacy of prompt engineering techniques in guiding LLMs to generate high-quality research survey papers, specifically exploring their application in the fields of artificial intelligence (AI) in drug development and large language model customization for classification. Using a mixed-methods approach, this research assessed the impact of prompt engineering on the coherence, relevance, and overall quality of generated content through quantitative and qualitative analyses. Techniques including prompt refinement, contextual prompting, and iterative prompting were systematically applied to enhance the …