Physical Sciences and Mathematics Commons^™

Understanding Deep Learning With Noisy Labels, Li Yi

Electronic Thesis and Dissertation Repository

Over the past decades, deep neural networks have achieved unprecedented success in image classification, which largely relies on the availability of correctly annotated large-scale datasets. However, collecting high-quality labels for large-scale datasets is expensive and time-consuming or even infeasible in practice. Approaches to addressing this issue include: acquiring labels from non-expert labelers, crowdsourcing-like platforms or other unreliable resources, where the label noise is inevitably involved. It becomes crucial to develop methods that are robust to label noise.

In this thesis, we study deep learning with noisy labels from two aspects. Specifically, the first part of this thesis, including two chapters, …

Copulas, Maximal Dependence, And Anomaly Detection In Bi-Variate Time Series, Ning Sun

Electronic Thesis and Dissertation Repository

This thesis focuses on discussing non-parametric estimators and their asymptotic behaviors for indices developed to characterize bi-variate time series. There are typically two types of indices depending on whether the distributional information is involved. For the indices containing the distributional information of the bivariate stationary time series, we particularly focus on the index called the tail order of maximal dependence (TOMD), which is an improvement of the tail order. For the indices without distributional information of the bivariate time series, we focus on an anomaly detection index for univariate input-output systems.

This thesis integrates three articles. The first article (Chapter …

Towards The Development Of A Cost-Effective Image-Sensing-Smart-Parking Systems (Isensmap), Aakriti Sharma

Electronic Thesis and Dissertation Repository

Finding parking in a busy city has been a major daily problem in today’s busy life. Researchers have proposed various parking spot detection systems to overcome the problem of spending a long time searching for a parking spot. These works include a wide variety of sensors to detect the presence of a vehicle in a parking spot. These approaches are expensive to implement and ineffective in extreme weather conditions in an outdoor parking environment. As a result, a cost-effective, dependable, and time-saving parking solution is much more desirable. In this thesis, we proposed and developed an image processing-based real-time parking-spot …

Towards A Novel And Intelligent E-Commerce Framework For Smart-Shopping Applications, Susmitha Hanumanthu

Electronic Thesis and Dissertation Repository

Nowadays, with the advancement of market digitalization accompanied by internet technologies, consumers can buy products from anywhere in the world. Finding the best-offered deal from numerous e-commerce sites and online stores is overwhelming, time-consuming, and often not very effective. Customers need to visit many online stores to find their desired product at the desired price. Also, the option of finding a product in the future time that is not currently available is limited in the current e-commerce platform. To address these limitations, there is a need to develop a new one-stop e-shopping model that would allow customers to search for …

Respiratory Pattern Analysis For Covid-19 Digital Screening Using Ai Techniques, Annita Tahsin Priyoti

Electronic Thesis and Dissertation Repository

Corona Virus (COVID-19) is a highly contagious respiratory disease that the World Health Organization (WHO) has declared a worldwide epidemic. This virus has spread worldwide, affecting various countries until now, causing millions of deaths globally. To tackle this public health crisis, medical professionals and researchers are working relentlessly, applying different techniques and methods. In terms of diagnosis, respiratory sound has been recognized as an indicator of one’s health condition. Our work is based on cough sound analysis. This study has included an in-depth analysis of the diagnosis of COVID-19 based on human cough sound. Based on cough audio samples from …

Improving Deep Entity Resolution By Constraints, Soudeh Nilforoushan

Electronic Thesis and Dissertation Repository

Entity resolutions the problem of finding duplicate data in a dataset and resolving possible differences and inconsistencies. ER is a long-standing data management and information retrieval problem and a core data integration and cleaning task. There are diverse solutions for ER that apply rule-based techniques, pairwise binary classification, clustering, and probabilistic inference, among other techniques. Deep learning (DL) has been extensively used for ER and has shown competitive performance compared to conventional ER solutions. The state-of-the-art (SOTA) ER solutions using DL are based on pairwise comparison and binary classification. They transform pairs of records into a latent space that can …

Automorphism-Preserving Color Substitutions On Profinite Graphs, Michal Cizek

Electronic Thesis and Dissertation Repository

Profinite groups are topological groups which are known to be Galois groups. Their free product was extensively studied by Luis Ribes and Pavel Zaleskii using the notion of a profinite graph and having profinite groups act freely on such graphs. This thesis explores a different approach to study profinite groups using profinite graphs and that is with the notion of automorphisms and colors. It contains a generalization to profinite graphs of the theorem of Frucht (1939) that shows that every finite group is a group of automorphisms of a finite connected graph, and establishes a profinite analog of the theorem …

Wideband And Relativistic Superradiance In Astrophysics, Christopher M. Wyenberg

Electronic Thesis and Dissertation Repository

In the quantum phenomenon of superradiance (SR) a population of inverted particles evolves, through its interaction with the quantized vacuum radiation field, into a highly entangled state capable of generating much greater radiative emission than predicted by the independent spontaneous decay of its constituent particles. The phenomenon has recently been applied to transient astrophysical processes but has thus far been restricted to particles sharing a common velocity. This thesis researches the effects of astrophysical velocity distributions upon SR, which are distinct from conventional regimes of the quantum optics literature in that they may possess extremely wide bandwidths, turbulent statistical properties, …

Analyzing Enhanced Electrochemiluminescence Of Novel Materials, Liuqing Yang

Electronic Thesis and Dissertation Repository

As a powerful analytical technique, electrochemiluminescence (ECL) has found wide applications in the fields of bioanalysis and light-emitting devices. However, many ECL materials sustain the shortcomings of low ECL efficiencies. To better design novel materials with ECL enhancement, it is essential to investigate and the enhanced ECL mechanisms. In this thesis, ECL enhancements of various novel materials were analyzed utilizing ECL-voltage curves, potential pulsing experiments, ECL and photoluminescence spectroscopies, and other characterization methods. Specifically, two graphene quantum dots (GQDs) with different heteroatom doping were designed and synthesized. Structural characterizations and spectroscopies demonstrated that the heteroatom doping could change the surface …

An Analysis Of Weighted Least Squares Monte Carlo, Xiaotian Zhu

Electronic Thesis and Dissertation Repository

Since Longstaff and Schwartz [2001] brought the amazing Regression-based Monte Carlo (LSMC) method in pricing American options, it has received heated discussion. Based on the research done by Fabozzi et al. [2017] that applies the heteroscedasticity correction method to LSMC, we further extend the study by introducing the methods from Park [1966] and Harvey [1976]. Our work shows that for a single stock American Call option modelled by GBM with two exercise opportunities, WLSMC or IRLSMC provides better estimates in continuation value than LSMC. However, they do not lead to better exercise decisions and hence have little to no effect …

Design And Surface Modification Of Noble Metal-Based Nanocatalysts, Xuchun Wang

Electronic Thesis and Dissertation Repository

This thesis investigates approaches to modifying the surface structure of noble metal-based nanocatalysts. Noble metal-based nanocatalysts, such as Pt and Pd, play a significant role in heterogeneous catalysis due to their capabilities in activating the cleavage or formation of chemical bonds, but still suffer from the high-cost issue and unsatisfied catalytic performance due to too strong or too weak adsorption of intermediates. Considering the surface specificity of heterogeneous catalysis, Bi, a cheap metal, was used to modify the surface of Pt- and Pd-based nanocatalysts. This thesis aims to unveil the role of Bi in improving their catalytic performance from both …

Geomechanical Modeling Of A Fault During Fluid Injection, Charles Kw Hulls

Electronic Thesis and Dissertation Repository

The injection of fluid into rock masses as a part of industrial processes, such as hydraulic fracturing, can lead to an increase in seismic activity. The movement of the injected fluid and resulting stresses can be simulated and analyzed. One aspect of this analysis is the predicted rate of seismic activity, obtained via the Dietrich rate-and-state law and the Coulomb Failure Stress. This work produces simulations for two fracturing scenarios in the Duvernay Shale region. Model parameters, such as layer permeability and timing of fault slip, are varied to determine their impact on the model results. The simulated results show …

Reduction Of L-Functions Of Elliptic Curves Modulo Integers, Félix Baril Boudreau

Electronic Thesis and Dissertation Repository

Let $\mathbb{F}_q$ be a finite field of size $q$, where $q$ is a power of a prime $p \geq 5$. Let $C$ be a smooth, proper, and geometrically connected curve over $\mathbb{F}_q$. Consider an elliptic curve $E$ over the function field $K$ of $C$ with nonconstant $j$-invariant. One can attach to $E$ its $L$-function $L(T,E/K)$, which is a generating function that contains information about the reduction types of $E$ at the different places of $K$. The $L$-function of $E/K$ was proven to be a polynomial in $\mathbb{Z}[T]$.

In 1985, Schoof devised an algorithm to compute the zeta function of an …

The Design And Implementation Of A High-Performance Polynomial System Solver, Alexander Brandt

Electronic Thesis and Dissertation Repository

This thesis examines the algorithmic and practical challenges of solving systems of polynomial equations. We discuss the design and implementation of triangular decomposition to solve polynomials systems exactly by means of symbolic computation.

Incremental triangular decomposition solves one equation from the input list of polynomials at a time. Each step may produce several different components (points, curves, surfaces, etc.) of the solution set. Independent components imply that the solving process may proceed on each component concurrently. This so-called component-level parallelism is a theoretical and practical challenge characterized by irregular parallelism. Parallelism is not an algorithmic property but rather a geometrical …

Reputation-Based Trust Assessment Of Transacting Service Components, Konstantinos Tsiounis

Electronic Thesis and Dissertation Repository

As Service-Oriented Systems rely for their operation on many different, and most often, distributed software components, a key issue that emerges is how one component can trust the services offered by another component. Here, the concept of trust is considered in the context of reputation systems and is viewed as a meta-requirement, that is, the level of belief a service requestor has that a service provider will provide the service in a way that meets the requestor’s expectations. We refer to the service offering components as service providers (SPs) and the service requesting components as service clients (SCs).

In this …

Characterization Of Some Natural Copper Samples From The Keweenaw Peninsula Of Lake Superior Using Electron Microscopy And X-Ray Tomography Techniques, Emilie Mk Landry

Electronic Thesis and Dissertation Repository

Copper corrosion is of interest to Canada’s Nuclear Waste Management Organization as it relates to their multi-barrier system for nuclear waste disposal in deep geologic repositories. Spent fuel canisters coated with a thin copper layer must persist for up to one million years. Natural copper from the Keweenaw Peninsula in Michigan, USA, has persisted for over 1 billion years and is here characterized to understand copper corrosion over geological time. Copper samples representing three rock types were characterized using optical microscopy, scanning electron microscopy techniques (energy dispersive X-ray spectroscopy and electron backscatter diffraction), and micro-computed X-ray tomography (microCT) to establish …

Developing Intelligent Routing Algorithm Over Sdn: Reusable Reinforcement Learning Approach, Wumian Wang

Electronic Thesis and Dissertation Repository

Traffic routing is vital for the proper functioning of the Internet. As users and network traffic increase, researchers try to develop adaptive and intelligent routing algorithms that can fulfill various QoS requirements. Reinforcement Learning (RL) based routing algorithms have shown better performance than traditional approaches. We developed a QoS-aware, reusable RL routing algorithm, RLSR-Routing over SDN. During the learning process, our algorithm ensures loop-free path exploration. While finding the path for one traffic demand (a source destination pair with certain amount of traffic), RLSR-Routing learns the overall network QoS status, which can be used to speed up algorithm convergence when …

Developing Artificial Intelligence And Machine Learning To Support Primary Care Research And Practice, Jacqueline K. Kueper

Electronic Thesis and Dissertation Repository

This thesis was motivated by the potential to use "everyday data", especially that collected in electronic health records (EHRs) as part of healthcare delivery, to improve primary care for clients facing complex clinical and/or social situations. Artificial intelligence (AI) techniques can identify patterns or make predictions with these data, producing information to learn about and inform care delivery. Our first objective was to understand and critique the body of literature on AI and primary care. This was achieved through a scoping review wherein we found the field was at an early stage of maturity, primarily focused on clinical decision support …

The Role Of Nonideal Magnetohydrodynamic Effects, Gravitational Instability, And Episodic Accretion In Star-Formation, Indrani Das

Electronic Thesis and Dissertation Repository

My dissertation focuses on the effect of magnetic fields on disk and core evolution during star-formation. We investigate the fragmentation scales of gravitational instability of a rotationally-supported self-gravitating protostellar disk using linear perturbation analysis in the presence of two nonideal magnetohydrodynamic (MHD) effects: Ohmic dissipation and ambipolar diffusion. Our results show that molecular clouds exhibit a preferred lengthscale for collapse that depends on mass-to-flux ratio, magnetic diffusivities, and the Toomre-Q parameter. In addition, the influence of the magnetic field on the preferred mass for collapse leads to a modified threshold for the fragmentation mass, as opposed to a Jeans mass, …

Complex Unicellular Microfossils From The 1.9 Ga Gunflint Chert, Canada, Ana L. González Flores

Electronic Thesis and Dissertation Repository

The presence of eukaryotic life during the early Paleoproterozoic has been a matter of debate because well-preserved fossils older than 1.8 Ga rarely exhibit eukaryotic cellular microstructures. In this study, microfossils from the 1.9 Ga Gunflint Chert were studied using the extended-focal-depth imaging technique, combined with scanning electron microscopy, resulting in recognition of three types of large (10–35 μm diameter) complex unicellular bodies (CUBs) and one type of “multicellular body” (< 50 μm diameter). The CUBs show the following eukaryotic cyst-like structures: (1) radially arranged internal strands similar to those in some acritarchs and dinoflagellates; (2) regularly spaced long tubular processes, stubby pustules, and/or robust podia on the cell surface; (3) reticulate cell-wall sculpturing such as pits, ridges, and scale-like ornaments; and (4) internal bodies that may represent membrane-bounded organelles. These morphological features provide strong evidence for the presence of protists in the late Paleoproterozoic.

Among the three types of CUBs from the Gunflint microbiota, a new species, Germinosphaera gunflinta sp. nov., was recognized. This species has the diagnostic characteristics of Germinosphaera, such …

Oxygen Speciation In Potassium Silicate And Potassium Aluminosilicate Glasses: Insights From X-Ray Photoelectron Spectroscopy, Ryan Sawyer

Electronic Thesis and Dissertation Repository

In a geological context, glasses are useful analogues for silicate melts as they are more readily studied in the laboratory using a wide range of techniques that are impractical for molten liquids. Understanding the structure of binary silicate glasses can help us understand more about the magmatic processes that affect terrestrial planetary bodies.

Potassium silicate glasses ranging in composition from 10 mol% to 35 mol% K₂O were studied using X-ray Photoelectron Spectroscopy (XPS). From high resolution O 1s XPS spectra, Bridging Oxygen (BO) mole fractions were calculated and compared with those of previous ²⁹Si MAS NMR studies. …

Examining Κ, The High Frequency Spectral Decay Parameter, In Eastern Canada, Samantha M. Palmer

Electronic Thesis and Dissertation Repository

This thesis examines the ground motion modeling parameter kappa (κ) in the stable continental region of eastern Canada. Kappa characterizes the decay of spectral amplitudes at high frequencies due to near-surface material de-amplification and is important in seismic hazard assessments. Kappa has significant economic and seismic safety implications for critical infrastructure such as nuclear power plants and dams.

To examine kappa in eastern Canada, a database of ground motions to analyze near-source ground motion characteristics is developed. The database consists of ground motion records from 3357 earthquakes of moment magnitude (M) ≥1.5 recorded within 150 km of 25 …

New Developments On The Estimability And The Estimation Of Phase-Type Actuarial Models, Cong Nie

Electronic Thesis and Dissertation Repository

This thesis studies the estimability and the estimation methods for two models based on Markov processes: the phase-type aging model (PTAM), which models the human aging process, and the discrete multivariate phase-type model (DMPTM), which can be used to model multivariate insurance claim processes.

The principal contributions of this thesis can be categorized into two areas. First, an objective measure of estimability is proposed to quantify estimability in the context of statistical models. Existing methods for assessing estimability require the subjective specification of thresholds, which potentially limits their usefulness. Unlike these methods, the proposed measure of estimability is objective. In …

Investigating The Influence Of Synthetic Environments On The Electronic Structure And Luminescence Of Cr-Doped Zinc Gallate, Xincheng Li

Electronic Thesis and Dissertation Repository

Cr doped ZnGa₂O₄ (Cr-ZGO) has shown potential in the field of bioimaging. To achieve the nano-size suiting for bioimaging, Cr-ZGO is normally synthesized using a hydrothermal method. There are few studies on how the hydrothermal conditions influence the structure and optical properties of the Cr-ZGO. The first two parts of this thesis study how different hydrothermal temperatures and pH environments will affect the optical properties of Cr-ZGO by photoluminescence spectroscopy. By combining the electronic structure information extracted from X-ray absorption near-edge structures (XANES) and extended X-ray absorption fine structure (EXAFS), a relationship among hydrothermal conditions, optical properties …

Magnetic Resonance Systems Development For Point-Of-Care Mri Platforms, Eric J. Lessard

Electronic Thesis and Dissertation Repository

Magnetic resonance imaging utilizes electromagnets to produce anatomical images in both clinical and research settings. In the race towards increasing performance head-optimized scanners have begun playing a significant role in providing high quality imaging of the head. However, they are implemented using smaller geometries and as such fail to allow entrance of the patient past their shoulders. This is overcome by designing asymmetric gradient coils which have their imaging region located towards one end of the gradient coil, as opposed to the geometric center, allowing brain imaging. There exists interest in compact configurations which allow imaging further into the cervical …

White Micas As A Tool For Tracking Pegmatite Evolution And Its Use In Li Exploration. A Case Study Of Wekusko Lake, Manitoba, Canada., David N. Benn

Electronic Thesis and Dissertation Repository

The Wekusko Lake pegmatite field is part of the Green Bay Group of pegmatites located near Snow Lake, Manitoba, Canada. Within this field, there are at least 13 spodumene-bearing pegmatites of varying degrees of mineralization. The abundance of white micas and the chemical affinity to incorporate Li and other trace elements into its crystalline structure makes white micas an ideal indicator mineral. The evolution of a melt can be tracked using trace elements in white mica. The K/Rb ratio vs Cs is used to determine a link between increasing evolution and increased Li contents of both the white mica and …

Probing The Inner Structure Of Active Galactic Nuclei Through Reverberation Mapping, Viraja Chandrashekhar Khatu

Electronic Thesis and Dissertation Repository

In the centres of massive galaxies, active galactic nuclei (AGN) are supermassive black holes, surrounded by an accretion disk of ionized gas, that release tremendous energy in the form of electromagnetic radiation. Because AGN are unresolved through telescopes, we employ reverberation mapping (RM) to study their structure. RM capitalizes on the fact that AGN are variable – the continuum emission from the accretion disk varies, and surrounding gas (in the broad-line region, BLR) responds to those variations with a positive time lag. RM translates the measured time lag into a size of the BLR. Combined with gas velocities (measured from …

Heat Flow In Terrestrial-Type Bodies From High P,T Electrical Resistivity Measurements Of Au, Fe-Si And Fe-Ni-Si Solid And Liquid Alloys, Meryem Berrada

Electronic Thesis and Dissertation Repository

The source of the fluid stirring mechanism that powers the dynamo of terrestrial-type bodies during their active magnetic field era is debated. Prior to the formation of a solid inner core, thermal convection may cause enough mechanical stirring of the core fluid to generate a magnetic field through dynamo action. After inner core formation, compositional convection in the liquid outer core becomes the main source of fluid stirring mechanism to power a dynamo. Constraints on the likelihood and duration of these convection mechanisms may be obtained by the experimental determination of the thermal properties of core materials. These cores consist …

Machine Learning With Big Data For Electrical Load Forecasting, Alexandra L'Heureux

Electronic Thesis and Dissertation Repository

Today, the amount of data collected is exploding at an unprecedented rate due to developments in Web technologies, social media, mobile and sensing devices and the internet of things (IoT). Data is gathered in every aspect of our lives: from financial information to smart home devices and everything in between. The driving force behind these extensive data collections is the promise of increased knowledge. Therefore, the potential of Big Data relies on our ability to extract value from these massive data sets. Machine learning is central to this quest because of its ability to learn from data and provide data-driven …

Deformation Conditions Of Quartz-Rich Mylonites Of The Grenville Front Tectonic Zone And Application To The Crustal Strength, Xueke Chang

Electronic Thesis and Dissertation Repository

This thesis analyzes a suite of mylonites from the Grenville Front shear zone exposed southeast of Sudbury, Ontario. Lattice preferred orientations, titanium-in-quartz thermometer, and dynamically recrystallized grain size piezometer measurements were applied to obtain the deformation mechanisms, deformation temperatures (T), and differential stresses (σ), respectively. Results show that these mylonites were formed in the shear zone during the terminal stage of the Grenville Orogeny. The dominant deformation mechanism is by regime 2 dislocation creep. The deformation temperature is between 425-567 °C, and the differential stress is between 56-133 MPa. These results are discussed in the context of wet quartzite flow …