Digital Commons Network^™

Investigation Of Dependency Parsing Techniques For Digital Document Analysis Through Deep Learning Approach, Rekah D Ms

Theses and Dissertations

Digital document dependency parsing is a significant task in natural language processing. Dependency parsing supports verification of the grammatical correctness of a sentence besides enabling extraction of relevant documents. Inappropriate extraction of features may result in falsely parsing a document, leading to decreased accuracy. Machine learning methods have been employed to perform feature extraction. However, selecting pertinent features was never achieved which minimizes the time consumption and overhead.

Hence, novel machine learning and deep learning techniques have been designed in our work for accurate and computationally efficient digital document analytics through dependency parsing. Four different contributions have been proposed for …

Bridging Scales In Epidemic Modeling From Individuals To Networks And Ecosystems, Saiprasasd Vr

Theses and Dissertations

Epidemics have shaped the trajectory of human history, repeatedly exposing vulnerabilities in healthcare systems and influencing socio-economic dynamics on a global scale. The COVID-19 pandemic, in particular, underscored the pressing need for flexible and responsive modeling framework capable of adapting to dynamic and evolving epidemic scenarios. Motivated by these challenges, this thesis explores the development of innovative epidemic models that integrate classical mathematical modeling techniques, network-based frameworks, and modern data-driven approaches. Each chapter addresses a distinct dimension of epidemic spread, from vaccination dynamics and community coupling to ecological complexity and adaptive human behavior, culminating in a model-free paradigm suitable for …

Cyclone Intensity Prediction In The Bay Of Bengal Using Deep Learning Methods, Senthil Kumar J

Theses and Dissertations

The Bay of Bengal region's coastlines have been badly devastated by tropical cyclones, as the region experiences an average of five to six cyclones per year, with about two to three of these intensifying into tropical storms or severe cyclones. Thus it necessitates to study the accurate and efficient forecasting of their intensity to improve preparedness and response to natural disasters. The present study compares and examines three distinct approaches to cyclone intensity prediction using historical datasets from 1998 to 2020: hybrid optimisation, deep learning-based, and empirical approaches.The predicted accuracy, computational effectiveness, and feasibility for real-time scenarios of each model …

Design Of An Integrated Lightweight Cryptographic Algorithm For Device Level Security And Fraternal Cryptographic Algorithm For Communication Security In Medical Cyber Physical Systems, Vimala Devi P

Theses and Dissertations

Healthcare involves detecting symptoms, diagnosing conditions and giving treatment to patients. It is one of the fundamental human rights, and it might be difficult to provide healthcare to those with chronic illnesses, elderly people with disabilities and those under distant observation. The World Health Organisation (WHO) states that Cardio Vascular Disease (CVD) is the leading cause of death worldwide.

According to the prediction, CVD-related causes such as heart attacks and strokes, could result in 23.3 million deaths by 2030. In addition, the number of people with diabetes will reach 246 million; thereby, the prevalence of CVD patients and diabetics will …

Investigations Of Secure Memory For Vlsi Based Crypto System, Vijay Sai R Mr

Theses and Dissertations

Semiconductor technology is growing very rapidly in their architectural developments, involving the usage of processor and memory. Presence of memory, in general is a vital commodity in various devices which are almost embedded into human activity, from robust work stations to handy mobile phones. Security of data stored in memory is very important, and hence, observation must be made that these valuable data should not be thwarted by malicious means. Security in cache memory is a major issue in memory related applications such as smart cards and bio-metric implementations.

Cache, is a small and limited memory located between central processing …

Evaluating Nfl Quarterback Decision Making Through Tracking Data, Jesse N. Melville

Theses and Dissertations

Quarterback decision-making is one of the most critical determinants of success in the National Football League (NFL), yet traditional statistics fail to account for the spatial, temporal, and contextual factors influencing each play. This thesis proposes a data-driven methodology for evaluating quarterback decisions using NFL player tracking data from the 2022 season. We replicate and extend the DeepQB framework by training deep learning models to estimate expected completion probabilities and expected yards gained for each receiver at the moment of the throw. Our extension applies the approach to a newer dataset with additional contextual features, enabling a more detailed evaluation …

A Dendrite Equivalence Relation On Loop Spaces, Spencer Arnesen

Theses and Dissertations

In Definition 2.0.3 we define an equivalence relation on paths such that two paths into a space X are related if doing the first, then the second backwards factors as a loop through a dendrite. This equivalence relation induces a group structure L(X) on a large subset of the loop space of X, namely all loops that have a unique reduced representative. We show that certain continuous maps induce a homomorphism between dentropy groups and that projection from three-dimensional real space to two-dimensional real space does not induce a homomorphism on dentropy groups.. We show that dentropy groups are locally …

Computational Modeling Of Magnetic Field Interaction With Superconductors Under High-Field And High-Dissipation Conditions: A Tdgl Approach Applied To Nb And Nb3sn, Aiden Victor Harbick

Theses and Dissertations

Superconducting radiofrequency (SRF) cavities are essential components in modern particle accelerators, enabling the efficient acceleration of charged particles for various applications in physics, medicine, materials science, and beyond. The performance of these cavities is significantly influenced by the properties of superconducting materials, such as niobium (Nb) and triniobium-tin (Nb3Sn), and the defects and surface features present within the material. This dissertation presents a computational study focused on understanding the behavior of SRF cavities, using a sample-specific time-dependent Ginzburg-Landau (TDGL) framework to simulate their performance under realistic material conditions. The research integrates experimental data and density functional theory (DFT) calculations to …

Mathematics Student Teachers' Curricular Reasoning During Planning, Erika Breann Miller

Theses and Dissertations

Teaching is a complex practice, in part because of all the decisions teachers make daily. This is especially true for student teachers who have little to no prior teaching experience to inform them about the decisions that need to be made in the classroom and how to reason through all those decisions as they plan and enact the lessons. Secondary mathematics student teachers were introduced to the Instructional Pyramid (Dingman et al., 2021) which organizes and maps teacher’s mathematical reasoning onto four elements of a classroom—teacher, student, mathematics, and curriculum—to examine which curricular reasoning aspects they use while planning mathematics …

Compact Heterogeneous Architectures And Algorithmic Methods For Enhanced Real-Time Digital True Time Delay Beamforming And Radar Applications, Nathan Lynn Burnett

Theses and Dissertations

Modern radar and signal processing systems require efficient computational architectures to handle increasing data rates while still being able to maintain dynamic configurability and complex tasks. This thesis presents and explores hardware and algorithmic approaches addressing these demands across several different applications. First, a low-cost tracking platform and architecture for unmanned aerial vehicles (UAVs) is developed, which provides detection and tracking capabilities, compliant with new FAA remote ID regulations. Second, a GPU-FPGA based digital true time delay beamformer is implemented to achieve real-time wideband radio frequency interference (RFI) mitigation. Specifically this beamformer explores weight calculation and calibration through particle swarm …

A Novel High-Order Absorbing Boundary Condition For Elastic Multiple Scattering Problems, Jordan Gregory Sheppard

Theses and Dissertations

Elastic waves are waves that propagate by causing deformations in an elastic medium such as metal, rock, or rubber. The main characteristic of elastic wave motion is the coexistence of multiple types of waves at once in the medium, including compressional and shear wave components. This type of motion can be used to study the impacts of physical phenomena such as earthquakes. In this paper, we discuss our research on the time-harmonic scattering of elastic waves in a medium by multiple circular cylindrical obstacles that the wave cannot penetrate. Using the Helmholtz decomposition, we break down the vector-valued physical system …

Versatile Waveform Generation And Clocking Techniques For Rfsoc-Based Digital Designs, Michael David Jones

Theses and Dissertations

Within an ever-evolving radio-frequency (RF) environment, designers are required to adapt systems to increasingly constricted parameters. The available tools continually grow to match this need, but so does the complexity in using them. The radio-frequency system-on-chip (RFSoC) is an innovation that combines a field-programmable gate array (FPGA) with an integrated RF chain. Designing with an RFSoC allows for high-precision data transmission, sampling, and processing, but it can be difficult and time-consuming to develop specialized circuits. The open-source CASPER project aims to provide a library of abstracted block-design components to simplify the design process. An arbitrary waveform generator (AWG) is created …

Topological Homotopy Groups, Tyler Evans

Theses and Dissertations

In general, the homotopy groups of topological spaces fail to be topological groups when endowed with the quotient topology. In this thesis, we show that for Peano continua with Abelian fundamental group, the fundamental group endowed with the quotient topology is indeed a topological group. However, this result does not extend to higher homotopy groups; that is, there exists a Peano continuum whose nth homotopy group, endowed with the quotient topology, is not a topological group. Additionally, a new characterization of the quotient topology in terms of a limit operator is introduced.

Development Of Numerical Tools And Performance Metrics For Analysis Of Optical Particle Counters, Doug Webb

Theses and Dissertations

Monitoring air quality is essential for understanding and mitigating the health impacts of airborne particulate matter (PM), which is linked to respiratory and cardiovascular disease. Optical particle counters (OPCs) are widely used for air quality monitoring, yet their precision and accuracy remain uncertain due to inconsistent performance evaluations. In this thesis, performance metrics are developed that can be used to assess the accuracy of OPCs in measuring PM concentrations and size distributions. Computational fluid dynamics (CFD) simulations are employed to analyze the behavior of particle-laden air streams through representative OPC models. The impact of design parameters is quantified using newly …

Tracing Tremor: From Upper-Limb Neuromusculoskeletal Dynamics To Movement Signatures In Essential Tremor, Landon J. Beutler

Theses and Dissertations

Tremor is a common neurological condition that significantly impairs the quality of life for millions. Despite this prevalence, the underlying mechanisms of its propagation remain poorly understood, and current treatment options often rely on trial-and-error methods. Moreover, the most common type, Essential Tremor (ET), presents a unique challenge due to its heterogeneity--patients exhibit a wide range of symptoms, severities, and responses to treatment. This thesis primarily advances tremor understanding through computational modeling, with a secondary focus on data-driven subtyping of ET via cluster analysis. These parallel paths aim to improve both mechanistic insight and clinical classification. Tremor propagation refers to …

Developing Site-Specific Water Response Functions Using Historical Soil Moisture Data, Gifty Lad Ayela

Theses and Dissertations

Variable rate irrigation (VRI) systems offer a promising solution by optimizing water application based on site-specific needs in different areas within a crop field. While accurate crop water production functions (CWPFs) are essential for VRI systems, developing CWPFs has been hindered by challenges in data collection and the complexities of modeling water’s impact on yields. This study attempts to develop site-specific CWPFs using farmers’ historical soil moisture data, collected using soil water tension (SWT) as a proxy, and yield maps. The spatial variability of the crop-water response functions was modeled using Geographically Weighted Regression (GWR) and the results from this …

Streamlined Technology Systems: Optimizing Technology Systems For Meridian Public School District, Latoya Atterberry

Theses and Dissertations

Technology has become a cornerstone of modern education, reshaping how teachers instruct and students learn. "Technology plays a pivotal role in shaping modern learning environments, offering opportunities for enhanced engagement and personalized learning experiences" (Ascione, 2023). This project focuses on the critical role of effective professional development in equipping teachers to integrate technology into their instructional practices meaningfully. Drawing on the TPACK (Technological Pedagogical Content Knowledge) and SAMR (Substitution, Augmentation, Modification, Redefinition) frameworks, this project examines how tailored professional development programs can bridge gaps in teachers’ technological and pedagogical knowledge, guiding them toward transformative technology use. This project culminates in …

Determining The Toxic Threshold And Other Use Characteristics For Industrial Hemp Seed Oil Used For Wood Product Protection, Daylan M. Barnes

Theses and Dissertations

Plant-based wood treatments have been utilized for many years, with varying success. Oils such as those extracted from linseed (flax), tung, and hemp plants have been used for thousands of years as finishes on wood. These are all drying oils, which dry to a film when exposed to air. Hemp seed oil has a high fatty acid content and can take a long time to dry on wood products but is valued in woodworking for being easy to use, food-safe, and having little to no odor. In this study, cold-pressed hemp seed oil, blended with a soybean carrier oil, is …

Calibration And Validation Of The Thermal Finite Element Model For Er120s-G Wire Arc Directed Energy Deposition Using Infrared Cameras And Thermocouples, Joshua Mykl Bassett

Theses and Dissertations

Wire Arc Directed Energy Deposition (WA-DED) is a promising additive manufacturing method for producing large metallic components with high deposition rates and low costs. However, its rapid heat input can lead to residual stress, microstructural changes, and dimensional inaccuracies. Accurate thermal modeling is essential to mitigate these effects. This study presents a thermal calibration of a finite element model (FEM) for ER120S-G using thermocouples and infrared (IR) imaging. Thermal data were gathered from three experiments to complete the calibration of the FEM . Calibration focused on tuning the Goldak heat source parameters, power efficiency, and heat transfer coefficients to minimize …

Autonomous Generation Of Ids Rules From Threat Intelligence, Azim Bazarov

Theses and Dissertations

Signature-based Intrusion Detection Systems (IDS) detect malicious activities by matching network or host activity against predefined rules. These rules are derived from Cyber Threat Intelligence (CTI), which includes attack signatures and behavioral patterns obtained through automated tools and manual threat analysis, such as sandboxing. The CTI is then transformed into actionable rules for the IDS engine, enabling real-time detection and prevention of threats. The constant evolution of cyber threats necessitates frequent rule updates, which delay deployment time and weaken overall security readiness. Recent advancements in autonomous agentic systems powered by Large Language Models (LLMs) offer the potential for automatic IDS …

Soil Ethics In Practice: Harvesting Conservation Choices Among Puerto Rican Coffee Farmers, Patricia Marie Cordero-Irizarry

Theses and Dissertations

Coffee farming in Puerto Rico has declined over recent decades, partly due to deteriorating soil quality. However, the decision-making process behind farmers’ adoption of soil conservation practices remains unclear, particularly the overlooked role of soil ethics, which emphasizes the moral principles guiding human relationships with the soil. This mixed-methods study examined coffee farmers’ adoption of soil conservation practices by integrating the Values-Beliefs-Norms (VBN) theory and a newly developed Soil Ethics Survey (SES) instrument for the quantitative component, while using Decision-Making Theory and Social Cognitive Theory to qualitatively explore how farmers’ experiences and reflections shape their soil ethics. Quantitative findings revealed …

Extreme Artifacts Removal With Curriculum Learning, Sushant Gautam

Theses and Dissertations

Restoring severely blurred images remains a significant challenge in computer vision, impacting applications in autonomous driving, medical imaging, and photography. This thesis introduces a novel training strategy based on curriculum learning to improve the robustness of deep learning models for extreme image deblurring. Unlike conventional approaches that train on only low to moderate blur levels, this method progressively increases the difficulty by introducing images with higher blur severity over time, allowing the model to adapt incrementally. Additionally, perceptual loss and hinge loss were integrated during training to enhance fine detail restoration and improve training stability. Various curriculum learning strategies were …

Understanding How Nws Meteorologists Tailor Hazardous Weather Messaging To Core Partners And Identify Local Vulnerabilities, Allison Camille Harvey

Theses and Dissertations

Over the past few decades, the National Weather Service (NWS) has continued to improve impact-based decision support services (IDSS) efforts to strengthen communication with core partners. This study addressed how a new spatially hazard-specific vulnerability tool called the Brief Vulnerability Overview Tool (BVOT) may impact how NWS meteorologists tailor messaging to their core partners. It was found that relationship building between NWS meteorologists and core partners is key to enhancing trust, especially during blue sky days. This was done by learning core partners’ needs, thresholds, and communication styles when presenting weather information. Core partners also noticed cues from NWS meteorologists …

Towards Applying Artificial Intelligence To Solve Np-Complete Problems Using Quantum Computing, Andrew Robert Haverly

Theses and Dissertations

This dissertation explores the methodology for more thoroughly entangling artificial intelligence and quantum computing. This is explored through a background search of problems solved using Grover’s quantum algorithm, a new quantum protein folding and drug discovery algorithm, using Grover’s algorithm to train quantum artificial neural networks, using quantum artificial neural networks for reinforcement learning, mapping classical assembly instructions to quantum circuits to make quantum programming easier, and using prompt engineering to get a classical artificial intelligence agent to solve an NP-Complete problem using a Grover’s algorithm. This method not only simplifies the creation of algorithms but also opens new avenues …

Topnet R1: A Multi-Stage Ai Framework For Topic Discovery In Scientific Abstracts, Md Elias Hossain

Theses and Dissertations

Scientific abstracts are rich sources of knowledge, yet extracting meaningful topics remains challenging due to limitations in existing topic modeling techniques. Traditional methods often struggle with interpretability, scalability, and contextual understanding. To overcome these issues, we introduce TopNet R1, a multi-stage ensemble framework that integrates traditional topic models with contextual embeddings from Bidirectional Encoder Representations from Transformers (BERT) and Generative Pre-trained Transformer 4 (GPT-4) large language models (LLMs). Top- Net R1 operates in three phases: (1) topic generation using Latent Dirichlet Allocation (LDA), Non-Negative Matrix Factorization (NMF), Latent Semantic Analysis (LSA), and Hierarchical Dirichlet Process (HDP); (2) LLM-assisted pattern recognition …

Spray Characterization And Fuel Property Analysis Of Tier 3 Gasoline In A Gdi System Using Machine Learning, Kyungwon Lee

Theses and Dissertations

This study investigates the spray behavior of Tier 3 gasoline in gasoline direct injection engines using experimental methods. Tier 3 gasoline, formulated with lower sulfur content, enhances aftertreatment system efficiency and supports cleaner combustion. A comparative fuel property analysis between Tier 3 gasoline and PACE-20, surrogate fuel, was conducted to understand compositional effects on spray. A constant volume spray chamber and optical diagnostics, including diffuse back-illumination extinction, Schlieren, microscopic, and 3D CT reconstruction imaging, were used to examine spray morphology under varying injection conditions. Key spray parameters such as liquid volume fraction, projected liquid volume, droplet size, and cone angle …

Exploring The Use Of A Random Forest In Correcting Gfs Tropical Cyclone Landfall Errors, Victoria Grace Maxwell

Theses and Dissertations

While Tropical Cyclone (TC) track forecasting has improved over the years it has been slow. These improvements can be attributed to increased computer capability and better model inputs. Machine learning has been used for forecasting different TC track and intensity. The study here looks to use a Random Forest to predict the TC latitude/longitude “landfall” at 72-hr and quantify improvements to the GFS forecast. Principal Component Analysis was used to reduce the dimensionality of the meteorological variables. Stepwise regression was used to determine the principal components that were important for forecasting landfall. These variables were then tested with a multivariate …

Unboxing The Black Box: Graph-Based Interpretability In Transformer Models, Ramak Nassiri

Theses and Dissertations

This thesis introduces a novel interpretability framework for transformer encoder models by hypothesizing that their internal embedding updates define a state transition system. We propose that transformers implicitly learn token-to-token influence dynamics, which can be analyzed using graph-theoretic methods. To validate this, we construct transition graphs from embedding changes and rank token importance using the PageRank algorithm. We develop two transformer models from scratch: a self-supervised model that incorporates serotype tokens to learn contextualized embeddings, and a supervised model initialized with these embeddings. Cosine similarity analysis of their token influence patterns reveals strong structural alignment, with values exceeding 93%. This …

Augmenting Healthcare Communication: Context-Aware Ai Frameworks For Clinical Decision Support And Automated Summarization, Subash Neupane

Theses and Dissertations

Healthcare communication is plagued by fragmented medical knowledge, patient misunderstanding of care plans, and clinician burnout from documentation burdens. These inefficiencies cost the U.S. healthcare system over $300 billion annually due to preventable nonadherence and administrative waste [47, 19]. While Artificial Intelligence (AI) technology like Large Language Models (LLMs) offer potential solutions, existing systems fail to deliver personalized, context-aware guidance or automate documentation without sacrificing accuracy. This dissertation addresses these gaps through three novel context-aware AI frameworks such as MedInsight, ClinicSum, and ClinicDuo. MedInsight leverages a multi-source context augmentation approach to synthesize patient centric medical responses by integrating Electronic Health …

Decision-Making And Execution Skill Estimation In Continuous Action Space Domains, Delma I. Nieves-Rivera

Theses and Dissertations

In many real-world domains with continuous action spaces, an agent’s performance is driven by two key factors: its decision-making skill, or ability to select high-quality actions, and its execution skill, or precision in carrying out these actions. However, accurately evaluating these skills, particularly when direct information about the agent’s intentions is unavailable, remains a challenging problem. This research addresses this need by developing a theoretical framework and proposing novel estimation methods to assess both decision-making and execution skills based solely on observable data: executed actions and their associated rewards. The work consists of six core studies focused on developing and …