Physical Sciences and Mathematics Commons^™

Robust Interventions In Network Epidemiology, Erik Weis

Graduate College Dissertations and Theses

Which individual should we vaccinate to minimize the spread of a disease? Designing optimal interventions of this kind can be formalized as an optimization problem on networks, in which we have to select a budgeted number of dynamically important nodes to receive treatment that optimizes a dynamical outcome. Describing this optimization problem requires specifying the network, a model of the dynamics, and an objective for the outcome of the dynamics. In real-world contexts, these inputs are vulnerable to misspecification---the network and dynamics must be inferred from data, and the decision-maker must operationalize some (potentially abstract) goal into a mathematical objective …

Effective Drag Coefficient Prediction On Single-View 2d Images Of Snowflakes, Cameron Hudson

Graduate College Dissertations and Theses

The drag coefficient of snowflakes is an crucial particle descriptor that can quantify the relationships with the mass, shape, size, and fall speed of snowflake particles. Previous studies has relied on estimating and improving empirical correlations for the drag coefficient of particles, utilizing 3D images from the Multi-Angled Snowflake Camera Database (MASCDB) to estimate snowflake properties such as mass, geometry, shape classification, and rimming degree. However, predictions of the drag coefficient with single-view 2D images of snowflakes has proven to be a challenging problem, primarily due to the lack of data and time-consuming, expensive methods used to estimate snowflake shape …

Employing Genetic Algorithms For Energy-Efficient Data Routing In Internet Of Things Networks, Farzana Akhter

Graduate College Dissertations and Theses

The Internet of Things (IoT) connects a vast number of smart objects for various applications,such as home automation, industrial control, and healthcare. The rapid advancement in wireless technologies and miniature embedded devices has enabled IoT systems to be deployed in various environments. However, the performance of IoT devices is limited because of the imbalance of data traffic on different router nodes. Nodes that experience high data volume will have a higher energy depletion rate and, as a result, will reach the end of their life quicker than other routers that have less data traffic. Genetic Algorithms are a well-known technique …

A Computational Journey Through Conspiracy Theories: A Genealogical Approach, Mohsen Ghasemizade

Graduate College Dissertations and Theses

In an era where misinformation and conspiracy theories (CTs) proliferate, this study presents an approach to understanding and categorizing CTs through the development of a detailed `family tree'. By adopting different definitions, we explore CTs as efforts to explain events through the lens of hidden, malevolent forces, distinguishing between actual conspiracies and theoretical beliefs without empirical proof. Leveraging an analysis of 1769 articles from fact-checking websites and employing Keyphrase Extraction, we compiled a dataset that led to the identification of 769 unique conspiracies. A RoBERTa-based binary classifier was developed, achieving an F1 score of 87\%, to distinguish CTs from non-CT …

Data Science And Mathematical Modeling For Humanitarian Response, Ollin Demian Langle Chimal

Graduate College Dissertations and Theses

This dissertation focuses on the exploration of societal responses to crises, with a par-ticular interest in existing socio-economic disparities, using tools of data science and mathematical modeling. The scope of the research is comprised predominantly around the COVID-19 pandemic, incorporating an in-depth analysis across six middle-income countries; Brazil, Colombia, Indonesia, Mexico, Philippines, and South Africa. With GPS data of approximately three million users, we found significant differences in the adherence to stay-at-home guidelines, revealing a great contrast between individuals in high-wealth and low-wealth areas, showing the disparities of who were more able to stay at home without risking their livelihood. …

On The Construction Of More Lifelike Devices, Shawn Lawrence Beaulieu

Graduate College Dissertations and Theses

Devices which blur the distinction between the living and the inanimate are being reported with heightened frequency. But what forms of organization, and what modes of internal change and worldly interaction, are required for truly lifelike devices, rather than ones which abstractly mimic life in simulation? This thesis presents results from two publications which attempt to mitigate a problem known to limit the performance of artificial neural networks, called ``catastrophic forgetting"; and a third paper which tries to articulate a vision for the construction of more lifelike devices—whose form, function, and putative environment are not conclusively specified prior to their …

How To Analyze Parental Conversation Online: A Computational Stack For Studying Vaccine Hesitancy., Carter Willets Ward

Graduate College Dissertations and Theses

Despite national and international organizations such as the CDC and WHO recognizing the value of vaccines and their importance in addressing public health concerns, there has been a decline in coverage for even the most established vaccines over the past three years. The global COVID-19 pandemic has contributed to this decline via decreases in medical resource accessibility and an increase in vaccine hesitancy. Even before the COVID-19 pandemic, WHO had recognized vaccine hesitancy as one of the top ten threats to public health. In the present work, we introduce a background account of (1) vaccine hesitancy and (2) anti-vax activism, …

Effects Of Morphology On Genetic Assimilation Of Learned Behavior, Natalie L. Tolley

Graduate College Dissertations and Theses

The Baldwin effect is an evolutionary theory regarding the assimilation of ontogenetic changes into a population's genome via selection pressure to entrench beneficial phenotypes discovered through learning. In evolutionary computation, the incorporation of learning into non-embodied agents allows them to navigate otherwise rough fitness landscapes by allowing for local exploration at particular points in that landscape. Prior work investigating the specific mechanisms by which learned behavior is genetically assimilated is almost entirely limited to non-situated, non-embodied simulations such as bitstring manipulation. However, recent research has demonstrated that genetic assimilation can be observed in embodied agents. Learning more about the ways …

Group-Level Frameworks For Data Ethics, Privacy, Safety And Security In Digital Environments, Juniper Lovato

Graduate College Dissertations and Theses

In today's digital age, the widespread collection, utilization, and sharing of personal data are challenging our conventional beliefs about privacy and information security. This thesis will explore the boundaries of conventional privacy and security frameworks and investigate new methods to handle online privacy by integrating groups. Additionally, we will examine approaches to monitoring the types of information gathered on individuals to tackle transparency concerns in the data broker and data processor sector. We aim to challenge traditional notions of privacy and security to encourage innovative strategies for safeguarding them in our interconnected, dispersed digital environment.

This thesis uses a multi-disciplinary …

Analyzing The Impact Of Cultural Factors On Happiness Levels In Arabic Language Tweets, Parisa Suchdev

Graduate College Dissertations and Theses

Culture is a fundamental force shaping our view of the world. Filtered through the stories we share on social media, our collective behavior both reflects and amplifies cultural impacts. The present study seeks to describe the effect of cultural factors, such as religion, on happiness scores in Arabic language tweets from January 2010 to June 2023. Our methodology involves using present tools called Hedonometer (https://hedonometer.org/) to study happy and sad events and StoryWrangler (https://storywrangling.org/) to study the usage of keywords related to those events. Our findings reveal a notable pattern of Twitter happiness declining following the start of the Arab …

Estimating Particle Velocity From Dual-Camera Mixed Reality Video Images Using 3d Particle Tracking Velocimetry, Thomas Chivers

Graduate College Dissertations and Theses

Mixed reality (MR) systems integrate diverse sensors, allowing users to better visualize and quantify surrounding environmental processes. Some existing mixed reality headsets include synchronized front-facing cameras that, among other things, can be used to track naturally occurring tracer particles (such as dust or snowflakes) to estimate particle velocity field in real time. The current work presents a 3D particle tracking velocimetry (PTV) method for use with MR systems, which combines various monocular cues to match particles between corresponding stereo images. Binocular disparity is used to estimate particle distance from an observer. Individual particles are tracked through time and used to …

Leveraging A Machine Learning Based Predictive Framework To Study Brain-Phenotype Relationships, Sage Hahn

Graduate College Dissertations and Theses

An immense collective effort has been put towards the development of methods forquantifying brain activity and structure. In parallel, a similar effort has focused on collecting experimental data, resulting in ever-growing data banks of complex human in vivo neuroimaging data. Machine learning, a broad set of powerful and effective tools for identifying multivariate relationships in high-dimensional problem spaces, has proven to be a promising approach toward better understanding the relationships between the brain and different phenotypes of interest. However, applied machine learning within a predictive framework for the study of neuroimaging data introduces several domain-specific problems and considerations, leaving the …

Developing Muscle Synergy Functions For Remote Gait Analysis, Nicole Marie Donahue

Graduate College Dissertations and Theses

Digital medicine promises to improve healthcare and enable its delivery to rural and underserved communities. A key component of digital medicine is accurate and robust remote patient monitoring. For example, remote monitoring of biomechanical measures of limb impairment during daily life could allow near real-time tracking of rehabilitation progress and personalization of rehabilitation paradigms in those recovering from orthopedic surgery. Wearable sensors have long been suggested as a means for quantifying muscle and joint loading, which can provide a direct measure of limb impairment. However, current approaches either do not provide these measures or require unwieldy wearable sensor arrays and/or …

An Analysis Of A Linear Algebra Based Group Key Exchange Protocol, Annie Zhang

Graduate College Dissertations and Theses

Group key exchange protocols are used to establish session keys, which can then be used as encryption keys to set up secure channels of communication, between more than two parties simultaneously. Many different group key exchange protocols exist and require security proofs in order to determine the strength of the protocol and answer the following questions: does the protocol provide authentication, and if so, to what degree? Does the protocol provide key secrecy? In this thesis we examine a particular group key exchange protocol that we call the \textit{vector space projection protocol} as first described in “A Group Key Establishment …

On The Enhancement Of Penetrating Radar Target Location Accuracy With Visual-Inertial Slam, Joshua Girard

Graduate College Dissertations and Theses

This paper presents research concerning the use of visual-inertial Simultaneous Localization And Mapping (SLAM) algorithms to aid in Continuous Wave (CW) radar target mapping. SLAM is an established field in which radarhas been used to internally contribute to the localization algorithms. Instead, the application in this case is to use SLAM outputs to localize radar data and construct three-dimensional target maps which can be viewed in augmented reality. These methods are transferable to other types of radar units and sensors, but this paper presents the research showing how the methods can be applied to calculate depth efficiently with CW radar …

Gauge Against The Machine: Improving Representations Within Sociotechnical Instruments To Enrich Context And Identify Biases, Joshua Minot

Graduate College Dissertations and Theses

The proliferation of digital data across all areas of society has transformed our ability to hypothesize, study, and understand social systems.From this richness of data we have seen the development of innovative instruments to study---and make decisions with---the digital artifacts of the modern day. These developments build on advancements in computation, connectivity, analytical methodologies, and sociological theories. The sociotechnical instruments we have developed have been revolutionary to how we understand society and how we conduct business, but with these broad leaps comes ample room (and need) for more nuanced advancements. As with the development of any field, as the digital …

Enhancing Cybersecurity Of Power Systems Using Machine Learning, Fayha Almutairy

Graduate College Dissertations and Theses

The continuous and accelerated digitalization of industries and technologies has made most of our daily activities obtrusively depend on electricity. Consequently, reliable power system operation became the cornerstone of economic sustainability and technological development. Unfortunately, the grown dependency of modern power infrastructure on Information and Communication Technology (ICT) has increased the risks of cyber-attacks. According to the most recent statistics, the electrical power sector is one of the significant fields in the number of cyber-attacks per year. The most devious types of cyber-attacks target the power system state estimation. Realtime state estimation aims to filter out the noise of measurements …

Building A Learning Healthcare System: A Path To Optimizing Big Health Data To Inform Clinical Care Decisions, Danne Charlotte Emily Elbers

Graduate College Dissertations and Theses

The explosive growth of data and computing power of the last decades has had large impacts on a myriad of domains, not in the least on one of society’s most complex systems: healthcare. In this work, a version of the resulting Learning Healthcare System (LHS) is explored and elements of it have been implemented and are in use at the Department of Veterans’ Affairs today. After an overview of what a LHS is and what it could be once executed in its full form, the chapters will describe in detail some of the individual elements and how they address cogs …

Secure And Private Federated Learning At Large Scale, Timothy Stevens

Graduate College Dissertations and Theses

We present novel techniques to forward the goal of secure and private machine learning. The widespread use of machine learning poses a serious privacy risk to the data used to train models. Data owners are forced to trust that aggregators will keep their data secure, and that released models will maintain their privacy. The works presented in this thesis strive to solve both problems through secure multiparty computation and differential privacy based approaches. The novel FLDP protocol leverages the learning with errors (LWE) problem to mask model updates and implements an efficient secure aggregation protocol, which easily scales to large …

Modeling The Heterogeneous Temporal Dynamics Of Epidemics On Networks, Andrea Joan Allen

Graduate College Dissertations and Theses

Mathematical models of infectious disease are important tools for understanding large-scale patterns of how a disease spreads through a population. Predictions of trends from disease models help guide public health prevention and mitigation measures. Most simple disease models assume that the population is randomly mixed, but real-world populations exhibit heterogeneous patterns in the way people interact. These differences in population structure can be represented by networks. Networks can then be incorporated into disease models by using various interdisciplinary concepts and tools. Yet even network disease models often overlook that populations change over time. In this thesis, two models of infectious …

Loss Of Precision In Implementations Of The Toom-Cook Algorithm, Marcus Elia

Graduate College Dissertations and Theses

Historically, polynomial multiplication has required a quadratic number of operations. Several algorithms in the past century have improved upon this. In this work, we focus on the Toom-Cook algorithm. Devised by Toom in 1963, it is a family of algorithms parameterized by an integer, n. The algorithm multiplies two polynomials by recursively dividing them into smaller polynomials, multiplying many small polynomials, and interpolating to obtain the product. While it is no longer the asymptotically fastest method of multiplying, there is a range of intermediate degrees (typically less than 1000) where it performs the best.

Some applications, like quantum-resistant cryptosystems, require …

Arrangements Of The Inputs And Outputs In The Multi-Robot Continuous Control Problem, Sida Liu

Graduate College Dissertations and Theses

The Multi-Robot Continuous Control (MRCC) problem in Deep Reinforcement Learning requires a single neural controller (agent) to learn to control the behavior of multiple robot bodies. When learning to control a single robot body, sensors and motors are arbitrarily connected to the input and output layers of the neural controller, respectively, and this arrangement does not affect the learnability of target robot behaviors. If and how such arrangement can affect learnability in MRCC---when dealing with multiple robots with different body plans---is as of yet unknown.

In this thesis, I demonstrate the following: (1) A neural controller can control a small …

Establishing Behavioral Baselines For Computational Systems: Two Case Studies, John Henry Ring

Graduate College Dissertations and Theses

The behavior of modern systems lives in a complex landscape that is unique to its particular application. In this work we describe and analyze the behavior of two modern computational systems: a Linux server and the National Market System (NMS). Though this work is diverse in both the type and scale of system under study, it is unified through the design and implementation of computationally tractable quantitative metrics aimed at defining the state of behavior of these systems. Understanding the behavior of these systems allows us to ensure their desired operation. In the case of a server we need to …

Quantifying Language Changes Surrounding Mental Health On Twitter, Anne Marie Stupinski

Graduate College Dissertations and Theses

Mental health challenges are thought to afflict around 10% of the global population each year, with many going untreated due to stigma and limited access to services. Here, we explore trends in words and phrases related to mental health through a collection of 1- , 2-, and 3-grams parsed from a data stream of roughly 10% of all English tweets since 2012. We examine temporal dynamics of mental health language, finding that the popularity of the phrase ‘mental health’ increased by nearly two orders of magnitude between 2012 and 2018. We observe that mentions of ‘mental health’ spike annually and …

Language-Based Analysis Of Differential Privacy, Chukwunweike Abuah

Graduate College Dissertations and Theses

Differential privacy (Dwork, 2006; Dwork et al., 2006a) has achieved prominence over the past decade as a rigorous formal foundation upon which diverse tools and mechanisms for performing private data analysis can be built. The guarantee of differential privacy is that it protects privacy at the individual level: if the result of a differentially private query or operation on a dataset is publicly released, any individual present in that dataset can claim plausible deniability. This means that any participating individual can deny the presence of their information in the dataset based on the query result, because differentially private queries introduce …

Spillover, Dilution, And Coinfection: Understanding The Spread Of Disease Within Managed And Native Bee Communities., Phillip A. Burnham

Graduate College Dissertations and Theses

Maintaining healthy pollinator communities is vital both for ensuring food securityand ecological diversity. However, managed honeybees and wild bee communities are under threat from an array of stressors including habitat loss, global change, pesticide use, poor beekeeping, and various pests and pathogens. Pathogens have been shown to be spilling over from managed bees into wild bee populations and are known to adversely affect colony function as well as increase mortality. Understanding transmission mechanisms related to general dynamics in this system will not only benefit pollinator health, but also gives us insight into important and understudied topics in disease ecology. In …

Quantifying Proverb Dynamics In Books, News Articles, And Tweets, Ethan Davis

Graduate College Dissertations and Theses

Proverbs are an essential component of language and culture, and though much attention has been paid to their history and currency, there has been comparatively little quantitative work on the frequency with which they are used, and the dynamics of their use over time. With wider availability of large corpora reflecting many diverse genres of documents, it is now possible to take a wider view of the importance of the proverb. Can a corpus linguistic approach to phraseology support existing histories, and what further insight can be gained from a quantitative approach? This study measures temporal changes in the relevance …

Perils And Pitfalls Of Symbolic Regression, Ryan Grindle

Graduate College Dissertations and Theses

The ever-growing accumulation of data makes automated distillation of understandable models from that data ever-more desirable. Deriving equations directly from data using symbolic regression, as performed by genetic programming, continues its appeal due to its algorithmic simplicity and lack of assumptions about equation form. However, few models besides a sequence-to-sequence approach to symbolic regression, introduced in 2020 that we call y2eq, have been shown capable of transfer learning: the ability to rapidly distill equations successfully on new data from a previously unseen domain, due to experience performing this distillation on other domains. In order to improve this model, it is …

Developing Natural Language Processing Instruments To Study Sociotechnical Systems, Thayer Alshaabi

Graduate College Dissertations and Theses

Identifying temporal linguistic patterns and tracing social amplification across communities has always been vital to understanding modern sociotechnical systems. Now, well into the age of information technology, the growing digitization of text archives powered by machine learning systems has enabled an enormous number of interdisciplinary studies to examine the coevolution of language and culture. However, most research in that domain investigates formal textual records, such as books and newspapers. In this work, I argue that the study of conversational text derived from social media is just as important. I present four case studies to identify and investigate societal developments in …

Exploring Hidden Networks Yields Important Insights In Disparate Fields Of Study, Laurence Clarfeld

Graduate College Dissertations and Theses

Network science captures a broad range of problems related to things (nodes) and relationships between them (edges). This dissertation explores real-world network problems in disparate domain applications where exploring less obvious "hidden networks" reveals important dynamics of the original network.

The power grid is an explicit network of buses (e.g., generators) connected by branches (e.g., transmission lines). In rare cases, if k branches (a k-set) fail simultaneously, a cascading blackout may ensue; we refer to such k-sets as "defective". We calculate system risk of cascading failure due to defective 2-sets and 3-sets in synthetic test cases of the Polish and …