Physical Sciences and Mathematics Commons^™

Predictive Inference For Ion Concentration With Machine Learning And Bayesian Methods, Alexandra B. Ulbing

Theses and Dissertations

Ultraviolet--visible (UV--Vis) spectroscopy produces high-dimensional signals that are strongly collinear, shift with concentration, and exhibit heteroskedastic, non-Gaussian noise. These features make supervised regression from spectra to ionic concentrations statistically challenging and limit the reliability of methods that assume linear structure or homoscedastic errors.

This dissertation develops two complementary frameworks for prediction and uncertainty quantification in UV--Vis spectroscopic regression: (1) frequentist stacked ensembles combined with distribution-free conformal prediction, and (2) Bayesian hierarchical modeling and Bayesian stacking. Together, they provide a unified view of model-based and distribution-free uncertainty across nickel and nickel--cobalt datasets.

The frequentist component builds ensembles of Functional Data Analysis …

Relationship Extraction Using Retrieval Augmented Generation For Biomedical Dataset, Jannat Jannat

Theses and Dissertations

With the increasing number of structured and unstructured data, obtaining reliable information effectively has become crucial. In the biomedical domain, extracting information from the scientific papers is crucial in order to stay up-to-date with accurate information, given the increased pace by which new research studies are published. This work focuses on identifying relationships between entities that are extracted from the abstracts and titles of biomedical research papers. In this work, we developed a Retrieval Augmented Generation (RAG) based system to automatically identify relations between biomedical entities. We evaluate multiple open source Large Language Models (LLMs) and the number of examples …

Cost-Effective Strategies For Feral Swine Control: Exploring Trapping Equipment Cooperatives, Phil Kenkel, Riza Radmehr, Rodney Holcomb, Mckenzie Boyce

Human–Wildlife Interactions

Feral swine (Sus scrofa) in the United States have become a growing concern, with a population of ≥6 million and causing annual damages of ≥$1.5 billion USD. Because of their high reproductive capacity, effective control methods are crucial to reducing or slowing their population growth. While remotely monitored and triggered traps have proven to be an effective control strategy, the cost of sophisticated trapping equipment, which can cost ≥$7,000 USD, is often cost prohibitive for many landowners. Despite the pressing need to address the challenge of making effective feral swine control methods more affordable, there are few studies …

Characterization And Optimization Of Sand And Tung Oil-Based Resins For Binder-Jet 3d Printing, Daniel I. Ajiola

College of Graduate Studies: Theses & Dissertations

Binder-jet 3D printing as a transformative technology in additive manufacturing, offers the ability to fabricate complex structures with diverse materials. This thesis investigates the use of a sustainable tung oil-based resin to create composites, exploring the potential for an eco-friendly alternative to synthetic binders.

The aim of this research is to develop and characterize a bio-based resin formulation, using tung oil as the primary binder, for application in binder-jet 3D printing with sand as the reinforcement. The resin formulation was prepared by combining tung oil, n-butyl methacrylate, divinylbenzene, and di-tert-butyl peroxide in precise proportions, ensuring a balanced mixture that supports …

Application Of Machine Learning And Large Language Models In Healthcare For Data Prediction And Summarization, Chiazam Chisom Izuchukwu

College of Graduate Studies: Theses & Dissertations

This study aims to examine the use of machine learning (ML) and large language models (LLMs) in healthcare to enhance disease prediction, clinical decision-making, and information management. Five supervised ML models—Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), Decision Trees (DT), and Naïve Bayes (NB)—on three different computing platforms—Google Colab, Databricks, and Snowflake—were employed for disease classification. Data preprocessing included treating missing values, encoding categorical variables utilizing one-hot-encoding, feature scaling when needed, and tackling class imbalance with Synthetic Minority Over-sampling Technique (SMOTE) before an 80-20 train-test separation. Models were created with Scikit-learn (Google Collab), Spark MLlib (Databricks), and …

Machine Learning Methods For Intrusion Detection And Response In Network Security, Ayomide Oyemaja

College of Graduate Studies: Theses & Dissertations

Intrusion Detection Systems (IDS) play a crucial role in computer network security by identifying malicious activities and potential cyberattacks. This thesis combines machine learning and cybersecurity by applying Reinforcement Learning (RL) in intrusion detection and response using the NSL-KDD dataset.

We designed and implemented a Q-learning framework where an agent learns to classify network traffic over time by interacting with the environment and receiving rewards based on detection accuracy. We also look at the importance of feature selection and classification techniques and how effective they are in improving model performance, reducing the complexity of computation, and producing more desirable results. …

Synthesis And Characterization Of A Biobased Polymer From Tung Oil And Pine Rosin & Characterization Of Polyhydroxybutyrate., Arijanet P. Osumah

College of Graduate Studies: Theses & Dissertations

This thesis presents two independents but thematically aligned research projects focused on the development and evaluation of sustainable polymers. The first involves the synthesis and characterization of a biobased polymer derived from tung oil and pine rosin. Tung oil, which is a triglyceride was chemically modified to produce tung oil diglyceride (TDG), which subsequently reacted with abietic acid-rich pine rosin to yield tung oil abietate (TOA), a biobased monomer. The TOA monomer was polymerized with butyl methacrylate (BMA) and divinylbenzene (DVB) using di-tert-butyl peroxide (DTBP) as a thermal initiator. The resulting polymer was thoroughly characterized using Fourier transform infrared spectroscopy …

Dynamic Analysis Of Malware Detection Using Customized Payloads: Examining The Effectiveness Of Manual And Automated Approaches In Web Applications, Jiban Krisna Das

College of Graduate Studies: Theses & Dissertations

Web applications are becoming the prime targets for cyber-attacks, where SQL injection (SQLi) and Cross Site Scripting (XSS) are the most exploited vulnerabilities. The study explores a novel approach using customized payloads to examine the effectiveness of manual and automated techniques of malware detection. This dynamic approach can effectively generate attack payloads and identify the vulnerabilities in a website thereby strengthening website security measures. This research focuses on dynamic analysis in a controlled environment while testing and analyzing SQL and XSS payloads under varying security conditions. This quantitative analysis involves crafting targeted payloads to bypass Web Application Firewall (WAF) filters …

Application Of Semantic Segmentation To An Automatic Target Recognition Problem, Bruce W. Rush Jr.

College of Graduate Studies: Theses & Dissertations

Synthetic Aperture Radar (SAR) is an active remote sensing system commonly used in aerial reconnaissance. SAR penetrates cloud cover and vegetation by recording reflected energy pulses. The resulting information content is difficult to interpret due to vast clutter data and sparse target data. The time-consuming process of analyzing SAR imagery can be greatly reduced by implementing an Automatic Target Recognition (ATR) algorithm. Convolutional Neural Networks (CNN) are capable of extracting identification features from SAR data content. Further research in this field indicates that high performing models are insufficiently robust due to high clutter correlation among classes in the Moving and …

Impact Of Y-Mixer Geometry And Flow Rate Modulation On Chemical Gradient Generators In Microfluidic Devices, Elizabeth Thurston Hutton

College of Graduate Studies: Theses & Dissertations

Various microfluidic structures are devised to generate tunable, stable spatiotemporal chemical gradients for single-cell studies. This work addresses the role of Y-mixer junctions in generating nonlinear gradients by increasing the width of one arm (bias), which is suspected to cause premature diffusion within the mixer. This study introduced a cinching feature to narrow the wider arm locally and redefine the junctions of these biased Y-mixers using four novel geometries defined by a 5-micrometer drawing rule. Furthermore, three flow rate factors allowed us to explore the impact on the generated gradient landscapes. The results showed that the four novel designs similarly …

Response Of Galactic Disks To Dark Matter Halo Triaxiality, Alex J. Hutton

College of Graduate Studies: Theses & Dissertations

In this study we examine the influence of dark matter halo triaxiality on the evolution of galactic disks. We do this by running a series of collisionless N-body models. In a first set of models we place the galactic disk within an oblate halo at various orientations and examine the resulting evolution. In our second set we construct fully self-consistent models of a stellar disk embedded within a spherical halo, with an impacting external dark matter dwarf galaxy. After evolving the simulations for a Hubble time, we examine the disk orientation and warping along with other quantities such as bar …

Majority Decision Using Top-Performing Neural Networks Models For Improved Credit Risk Prediction, Vincent Dey

College of Graduate Studies: Theses & Dissertations

Credit risk prediction remains both a challenging and high-interest problem due to the inherently unbalanced nature of financial datasets and the continuous drive for higher pre- dictive precision. In this work, I build upon previous advancements in credit risk modeling and introduce an ensemble-based Artificial Neural Network (ANN) architecture designed to enhance classification performance. By leveraging a selective ensemble of decision net- works, this approach not only improves prediction accuracy but also mitigates the chal- lenges posed by imbalanced data distributions. While the primary focus is on credit risk prediction, my analysis demonstrates that the proposed model can be effectively …

Applying The Lagrangian Variational Method To Atom Interferometry, Jeffrey W. Heward

College of Graduate Studies: Theses & Dissertations

Atom interferometry in Bose-Einstein condensate systems has emerged as a promising technique for precision metrology. The dynamics of these systems are well-described by the Gross-Pitaevskii equation (GPE), but direct numerical solution of this equation is infeasible for realistic systems. We use the Lagrangian Variational Method (LVM) to approximate solutions of the GPE in 1D and 3D, and we compare the LVM results to the exact solutions. We also present 3D LVM results for a case where numerical solution of the GPE is infeasible.

Towards A Baker-Campbell-Hausdorff Theorem In Positive Characteristic, Nikolas A. Koutroulakis

College of Graduate Studies: Theses & Dissertations

We investigate whether there is an analog of the Baker-Campbell-Hausdorff (BCH) theorem for Lie algebras over fields of positive characteristic. We begin by introducing the proof of the BCH formula in characteristic zero. We then introduce the Artin-Hasse exponential and show that it is $p$-integral. Our main result provides sufficient conditions under which a BCH-type formula exists for the Artin-Hasse exponential in positive characteristic. Additionally, we derive a formula for computing an inverse of the Artin-Hasse exponential.

Assessment And Health Benefit Of Vitamin C Using A549 Cell, Yetunde Adepoju

College of Graduate Studies: Theses & Dissertations

Vitamin C (ascorbic acid) is a vital micronutrient recognized for its antioxidant capacity and essential bodily functions, such as collagen formation, neurotransmitter activity, and regulation of the immune system. This study explores how vitamin C can protect and potentially heal cells under oxidative stress and inflammation, using A549 cells, a human lung epithelial cell line, as the model system. The cells were exposed to varying concentrations of vitamin C dissolved in water, DMSO, and 5% DMSO. To assess the outcome, cell viability was measured with an MTT assay. In addition, the expression of key genes related to oxidative stress and …

Seafloor Magnetic Alteration Through Time Across The Northern Mid Atlantic Ridge, Jelil Olamide Olaoye Mr

College of Graduate Studies: Theses & Dissertations

The magnetic properties of oceanic crust evolve with age due to chemical alteration, changes in magnetic mineralogy, and physical processes associated with seafloor spreading. This study investigates how magnetic properties change over time in basaltic rocks from the northern Mid Atlantic Ridge, with a focus on two V-shaped ridge sites (VSRs; U1562 and U1563), two V-shaped troughs (VSTs; U1554 and U1555) and a normal crust site formed during episodic mantle upwelling. Magnetic analyses including anisotropy of magnetic susceptibility (AMS), natural remanent magnetization (NRM), alternating field demagnetization (AF) and hysteresis loops, was performed with optical microscopy and scanning electron microscopy (SEM) …

Facile Synthesis Of Gold, Copper, And Silicon Nanostructures By Laser Ablation In Liquid, Nicholas Simpson

Theses and Dissertations

Metal nanoparticle (NP) synthesis is a continually evolving area of research, owing to the widespread application of NPs in fields such as catalysis, drug delivery, and biosensing. Compared to bulk metal material, NPs contain a larger proportion of surface atoms and exhibit unique optical properties, thus making them ideal for applications where maximizing surface interactions is key. Moreover, their compositions can be altered in several different ways, as evidenced by metal alloy, core/shell, supported, and defective NPs. While conventional synthesis methods are capable of producing NPs in large quantities, they often require conditions of high temperature and pressure, strong reducing …

Dosimetric Validation Of An Electron Monte Carlo (Emc) Algorithm For Small Cutouts And Extended Ssd’S: Assessing Emc In Small Field Dosimetry, Anil Basavaaraju

Theses and Dissertations

This thesis validates the accuracy and limitations of Electron Monte Carlo (eMC) simulations in Eclipse treatment planning systems for small-field electron dosimetry and extended source-to-surface distances (SSDs). Systematic measurements were performed using a microDiamond detector across various electron energies and field sizes. The study examines the discrepancies between measured and calculated (eMC) dose distributions, validating challenges due to lateral scatter disequilibrium and extended air gaps. Our findings provide systematic data for eMC dose calculations and demonstrate that eMC performs within clinical tolerances, with the ultimate goal of enhancing the precision and safety of radiation treatments.

Multiphysics Modeling Of Material Response To High-Intensity X-Ray And Laser Pulses: Heating, Ablation, And Plasma Expansion, Youssef Abouhussien

Theses and Dissertations

This dissertation presents a computational framework to investigate material response under high-intensity X-ray fluxes produced by an exo-atmospheric nuclear detonation and laser-material irradiations, with a focus on heating, ablation, and plasma expansion phenomena relevant to satellite vulnerability and high-energy-density environments. A hybrid Monte Carlo and Two-Temperature Model (MC-TTM) was developed to simulate X-ray and laser energy deposition and thermal relaxation in metals and semiconductors across a range of X-ray and laser pulse durations from femtoseconds to nanoseconds. Results demonstrate distinct thermal behavior between materials, with ablation thresholds and phase transitions captured in good agreement with experimental data.

In parallel, a …

Development, Synthesis And Characterization Of Advanced Low- Cobalt And High-Nickel Cathode Materials For Lithium-Ion Batteries, Arjun Patel

Theses and Dissertations

This dissertation focuses on the improvement of structural and electrochemical stability of active materials for lithium-ion batteries. It explores the modification strategies to mitigate the issues related to Ni-rich cathode materials, along with the continuous production of cathode precursors using a slug flow reactor to improve the quality of the product.

The First part of the dissertation examines the effect of substitution of cobalt with aluminum and iron on the electrochemical stability of Ni-rich lithium nickel cobalt manganese oxide (NCM) materials. This study investigates how Al and Fe doping work and how they impact the performance and stability of these …

Leveraging Distributed Semantics From Deep Learning Architectures For Literature-Based Discovery, Clint A. Cuffy

Theses and Dissertations

Literature-based discovery (LBD) is a scientific process that introduces methods to automatically identify novel insights between non-interacting sets of literature. To date, numerous statistical and machine learning-based methods have been applied in the biomedical domain to find treatments for diseases such as Raynaud's disease, Parkinson's disease, and Multiple Sclerosis. However, the lack of standardized practices and creation of bespoke methodologies produces a scenario where the adoption of LBD remains challenging in real-world systems. Our work addresses these concerns through the improvement of five critical areas: 1) error propagation within LBD's a priori dependent tasks, 2) exploring the integration of modern …

A New Robust Imputation Method For Longitudinal Data With Non-Normal Continuous Outcomes, Ahmed M. Gad Prof, Yasmie A. Mohamed, Nesma M. Daewish Dr, Abdelnaser S. Abdrabou Prof, Wafaa M. Ibrahim Dr

Business Administration

Missing values is very common in longitudinal data and it is the main challenge in analysis of longitudinal data. Missing values have a significant effect on longitudinal data analysis because they lead to loss of information, biased estimates, and misleading results. In practice there is a need for an imputation method to deal with missing values.

Aim: In this study a new robust regression-based imputation method to deal with missing values in longitudinal data is proposed. This method utilizes the modified adaptive linear regression model and does not require the normality of the responses. It is a novel robust imputation …

Comparative Study Of Single Imputation Techniques For The Prediction Of Missing Dairy Data, Ahmed M. Gad Prof, Ahmed Abdelhakim Ahmed Mr, Eman Manaa Prof, Basant Shafik Dr, Sakr Mostafa Prof

Business Administration

Dairy farm records are a crucial component of effective livestock business management. Record analysis allows a farm’s owner to make informed decisions. Incomplete records are less useful for data analysis, so it's important to handle missing values correctly. This study compares different imputation methods for handling missing values in a dataset of dairy records comprising 997 records collected from 234 cows between 2012 and 2022. The dataset was screened against records with missing values and then deleted, resulting in 858 observations from 200 animals. There were missing values in two variables, with a missing percentage of 13.9%: days in milk …

Towards Understanding Thermal Management In Unsteady Boundary Layer Flow With Ac/Dc Electric Fields, Sara Abdelsalam, M. A. Dagher, Y. A. Elmaboud, A. I. Abdellateef

Basic Science Engineering

Unsteady boundary layer flow induced by alternating current (AC) or direct current (DC) electric field through a porous layer is investigated numerically. The finite difference method based on Crank-Nicolson is applied to solve the nonlinear system. The governing equations are built with fractional shear stress and the Cattaneo heat flux model, and time fractional derivatives are computed using the Caputo fractional derivative. The numerical results are presented to demonstrate the effects of varying parameters on momentum and thermal boundary layer. The results reveal that the time delay in the velocity profile occurs for larger values of both the velocity fractional …

Semi Analytical Solution Ofmhd And Heat Transfer Of Couple Stress Fluid Over A Stretching Sheet With Radiation In Porousmedium, Sara Abdelsalam, M. Khairy, W. Abbas, A.M. Megahed, M.S. Emam

Basic Science Engineering

This comprehensive research examines the dynamics of magnetohydrodynamic (MHD) flow and heat transfer within a couple stress fluid. The investigation specifically focuses on the fluid’s behavior over a vertical stretching sheet embedded within a porous medium, providing valuable insights into the complex interactions between fluid mechanics, thermal transport, and magnetic fields. This study accounts for the significant impact of heat generation and thermal radiation, crucial factors for enhancing heat transfer efficiency in various industrial and technological contexts. The research employs mathematical techniques to simplify complex partial differential equations (PDEs) governing fluid flow and heat transfer. Specifically, suitable similarity transformations are …

A Finite Difference Analysis Of Electroosmotic Flow Of A Two-Layer Jeffrey Fluid In A Vertical Porous Nanochannel, Y. Abdelmaboud, M. A. Elkot, Sara Abdelsalam

Basic Science Engineering

In this paper, we discuss the unsteady two-layer electroosmotic flow (EOF) of the Jeffrey fluid in a vertical porous nanochannel, taking into account the effect of the heat transfer and alternating current (AC) electric fields. In this model, the two layers have different densities and viscosities. The finite difference method is used to solve the governing equations of electrical potential, velocity, and temperature. The results show that the Debye-Hückel parameter is significant in managing electroosmotic flow (EOF) because it affects the electrical double layer (EDL) compression and velocity gradients on the wall. The higher the radiation parameter, the lower the …

Numerical Simulation Via Homotopy Perturbation Approach Of A Dissipative Squeezed Carreau Fluid Flow Due To A Sensor Surface, Sara Abdelsalam, W. Abbas, A.M. Megahed, H. M.H. Sadek, M.S. Emam

Basic Science Engineering

This study rigorously examines the interplay between viscous dissipation, magnetic effects, and thermal radiation on the flow behavior of a non-Newtonian Carreau squeezed fluid passing by a sensor surface within a micro cantilever channel, aiming to deepen our understanding of heat transport processes in complex fluid dynamics scenarios. The primary objective is to elucidate how physical operational parameters influence both the velocity of fluid flow and its temperature distribution, utilizing a comprehensive numerical approach. Employing a combination of mathematical modeling techniques, including similarity transformation, this investigation transforms complex partial differential equations into more manageable ordinary ones, subsequently solving them using …

Investigations Of Conjugate Heat Transfer And Fluid Flow In Partitioned Porous Cavity Using Darcy-Forchheimer Model: Finite Element-Based Computations, Nasir Yasin, Shafee Ahmad, Muhammad Umair, Zahir Shah, Narcisa Vrinceanu, Ghadah Alhawael

Mathematics & Statistics Faculty Publications

The conjugate heat transfer and fluid flow has vast applications in thermal engineering, particularly for cooling in thermal devices, and automobile engines. This study investigates conjugate heat transfer in 2D enclosures, featuring thin solid fins attached to a porous bottom wall. The porous medium is considered isotropic and homogeneous by the Darcy-Forchheimer model, with fluid phases in local thermal equilibrium. The boundary conditions at the porous fluid interface ensure continuity of the velocities, stresses, temperature, and heat flux. The phenomenon is mathematically modelled by obtaining a set of partial differential equations. The finite element method (FEM) is used to perform …

Advancing Root Trait Measurements For A Deeper Understanding Of Soil Biogeochemistry, Brian Rinehart

Theses and Dissertations--Plant and Soil Sciences

Climate change is disrupting (agro)ecosystems, with the potential for significant losses in crop productivity and ecosystem resilience due to changes in water and biogeochemical cycling. These disruptions can also release more carbon dioxide into the atmosphere through the degradation of soil organic matter. Plant roots have emerged as a potential tool for mitigating these effects, given their roles in plant function and in driving soil biogeochemistry. While this interest is growing, our ability to implement root trait selection is limited due to constraints on characterizing roots and limited clarity on the effects of traits on plant and ecosystem function. My …

Giving Props To Soil Property Predictive Models: Utilizing Mir To Predict Soil Health Properties, Diala M. Abboud

Theses and Dissertations--Plant and Soil Sciences

Soil health is critical to sustaining life on Earth, influencing plant growth, agricultural productivity, water quality, and overall ecosystem sustainability. In response to climate change and intensifying land use, the need to improve our understanding of soil health has become increasingly urgent. The application of mid-infrared (MIR) spectroscopy coupled with statistical modeling has emerged as a promising, cost-effective approach to predict soil health parameters, offering a quicker alternative to traditional wet chemistry analyses. MIR methods have been successfully applied to predict various soil health properties, such as bulk density (BD), cation exchange capacity (CEC), base saturation (BS), electrical conductivity (EC), …