Physical Sciences and Mathematics Commons^™

Using Machine Learning Models To Discover Promising Research, Akhil Pandey Akella

Graduate Research Theses & Dissertations

Endeavors to identify valuable research involve the factors of discovery, comprehensibility, and reproducibility. The purpose of this study is to assist scholars in finding research that is both promising and of high quality. I explain how we can approach the problem of reproducibility in relation to scholarly articles and propose gauging the public understanding of science as a way to determine the comprehensibility of given research articles. Additionally, I explain how the concept of long-term social media impact supports the discovery of scholarly articles likely to be impactful even with the passage of time. I build and describe machine-learning models …

Sky Surveys Scheduling Using Reinforcement Learning, Andres Felipe Alba Hernandez

Graduate Research Theses & Dissertations

Modern cosmic sky surveys (e.g., CMB S4, DES, LSST) collect a complex diversity of astronomical objects. Each of class of objects presents different requirements for observation time and sensitivity. For determining the best sequence of exposures for mapping the sky systematically, conventional scheduling methods do not optimize the use of survey time and resources. Dynamic sky survey scheduling is an NP-hard problem that has been therefore treated primarily with heuristic methods. We present an alternative scheduling method based on reinforcement learning (RL) that aims to optimize the use of telescope resources for scheduling sky surveys.

We present an exploration of …

Improved Momentum Spread For Precision Physics Experiments Using Wedges, Nick Amato

Graduate Research Theses & Dissertations

High precision experiments have become key in the search for physics beyond the Standard Model. One measurement that can be made in order to expand our knowledge of the Standard Model is the precise measurement of the magnetic dipole moment (MDM) of the muon. In order to perform this precise measurement of the MDM of muons, the storage of particles with a precisely known momentum in a highly uniform magnetic field is required.

This thesis is focused on how one may use a wedge of energy absorbing material and the optical parameters of an accelerator lattice to improve upon the …

Morphometric Analysis Of Subfossil Macronycteris Spp. (Chiroptera: Hipposideridae) From Madagascar, Jamie Lynn Alumbaugh

Graduate Research Theses & Dissertations

Macronycteris bats are morphologically conservative between species but demonstrate intraspecific morphological variation between geographic locations and sexes. Two of the four living species of Macronycteris are found on Madagascar, where they are broadly distributed and demonstrate a trend in body size correlated with the latitudinal precipitation cline on the western side of the island. The presence of an extinct species, M. besaoaka, from Anhjohibe Cave in northern Madagascar suggests that Macronycteris was once more diverse, at least with respect to morphology. Since its description, taxonomic and phylogenetic revisions have reshaped our understanding of this genus. On Madagascar, these include the …

Possible Hydroseismic Triggering For Small Earthquakes Occurring Within The Fox River Valley, Northeastern Illinois, James Benco

Graduate Research Theses & Dissertations

In February, 2010 the town of Lily Lake experienced a magnitude 3.8 earthquake. This was followed by small earthquakes of magnitude 2-3 in the same general area in 2012 (McHenry), 2013 (Campton Hills) and 2015 (Lake of the Hills). Seismicity within this region is poorly understood. The Fox River Valley (here defined as the Fox River drainage basin) area is not historically seismically active, nor does it lie above a mapped fault. This study examines the timing of these earthquakes and the hypothesis they were triggered by pore-pressure increases at depth from large rainfall events several months earlier. A pressure …

Impact Of Agreeableness On Virtual Team Performance Through Team Identification And Shared Mental Models, Alexandria Brown

Graduate Research Theses & Dissertations

Virtual teams help organizations efficiently utilize their employees for a task without the requirement of co-location. The literature on team performance suggests that teamwork is integral to a team’s success; however, in virtual teams this is often a challenge. Certain personality characteristics on virtual teams may be particularly important to the development of effective teamwork. An under-investigated factor is the role agreeableness in virtual team processes and how it affects the overall team performance. The main research question of this study is how the degree of agreeableness on a virtual team affects the overall team performance through predicted associations with …

Utilizing Radiolarian Assemblages To Track Changes In The Agulhas Current Through The Mid-Pleistocene Transition: Iodp Site U1479, William Alberto Bugbee

Graduate Research Theses & Dissertations

The objective of this study was to assess the radiolarian assemblage of the Agulhas Current at International Ocean Discovery Program Expedition 361, Site U1479 (the most westward site), through the Mid-Pleistocene Transition (MPT). Samples between 1,000 to 777ka were utilized to (1) characterize the radiolarian assemblage, (2) observe whether the radiolarian assemblage is sufficiently distinctive to help discern the occurrence of the MPT in the Southern Ocean, and (3) determine whether changes in the radiolarian assemblage reflect a change or changes in ocean currents that may have occurred during the MPT.

Over 300 radiolarian species from 15 different families were …

Resonant Extraction For Mu2e, Prudhvi Raj Varma Chintalapati

Graduate Research Theses & Dissertations

Charged Lepton Flavor Violations (CLFVs) are probed in an attempt to search Beyond Standard Model (BSM) Physics. The sensitive channel available for experiments around the world is to probe through muons. No conclusive evidence of CLFV is found so far but $\mutoe$ experiment at Fermi National Accelerator Laboratory (Fermilab) aims to improve the current experimental sensitivity by searching for $\mu^{-}N \longrightarrow e^{-}N$ transition. To provide the high intensity beam that can be handled to achieve desired sensitivity, slow spill using resonant extraction is used at the accelerator complex.

Understanding the motivations behind the experiment and simulating the crucial stage in …

Maximum Likelihood Estimation For A Heavy-Tailed Mixture Distribution, Philippe Kponbogan Dovoedo

Graduate Research Theses & Dissertations

In an increasingly connected global environment, “high-impact, low-probability" (HILP)

events can have devastating consequences and result in large insurance losses with a heavy-

tailed distribution. Examples of such events include Hurricane Katrina, the Deepwater

Horizon oil disaster and the Japanese nuclear crisis and tsunami. According to the 2012

Blackett Review of HILP Risks from the UK Government Office for Science, the

identification of low-probability risks, and the subsequent development of mitigation plans,

is complicated by their rare or conjectural nature, and their potential for causing impacts

beyond everyday experience. Extremal mixture models and more generally extreme value

analysis help assess …

Study Of Higgs Production From H -> Zz -> 4l Channel Using Machine Learning Methods, Daniel Arthur Faia Jr.

Graduate Research Theses & Dissertations

In this thesis I will show how machine learning methods can improve on physics analysis in the H -> ZZ -> 4l channel. In particular we will explore how these methods can be used to classify Vector Boson Fusion (VBF) processes in the presence of more dominant Higgs production processes. The aim is to improve the ability to discriminate VBF Higgs boson production relative to other Higgs boson production modes. Since VBF has two quark jets in the final state, it is useful to discriminate between quark and gluon jets. We compare the effectiveness of quark gluon discrimination with machine …

A Review Of Reasons For Failure In Applying Machine Learning To Financial Trading And An Experiment Investigating Combinatorial Purged Cross Validation’S Merit In Preventing The Most Prominent Of These Reasons, Multiple Testing Bias, Colin Fritz

Graduate Research Theses & Dissertations

The interest in applying machine learning to financial trading in the hedge fund industry has exploded in the last five years due to the massive success of a handful of ‘quantitative’ investment firms like Renaissance Technologies who has pioneered the use of machine learning techniques in investment since the 1980s. The failure rate of such firms attempting to deploy financial machine learning strategies is very high. This thesis reviews many of the causes for failure such as harmful correlations between examples in the dataset, redundant observations, improper data sampling paradigm, and multiple testing bias. Of these, multiple testing bias is …

Textile Dyes As A Source Of Groundwater Contamination In Mandalay, Myanmar, Surya Freeman

Graduate Research Theses & Dissertations

Textile dyeing and weaving of traditional longyi garments has traditionally occurred in the Amarapura Township of Mandalay, Myanmar, since 1822, transitioning from natural to chemical dyes in the early 1900’s. With no current wastewater treatment facilities in Mandalay, dye effluents mix with other wastewaters in unlined canals dug near peoples’ homes and discharge into local canals and groundwater. As locals rely heavily on dug and tube wells for drinking, bathing, and cooking, this industry poses a major health hazard to the people in this region. The objective of this study is to identify the previously unknown composition of the textile …

Impact Of Standardized Test Performance On Success In Introductory College Physics Classes, Michael Gattone

Graduate Research Theses & Dissertations

This research project looks to analyze the ability of the ACT and other standardized tests to predict student success in introductory physics classes at Northern Illinois University (NIU). Admissions data from NIU was collected by the physics department from a total of 4262 students enrolled in introductory physics courses during the Fall 2012 semester until the Spring 2017 semester. This data includes academic level, gender, race/ethnicity, composite ACT score, high school GPA, and final letter grade. In particular, the composite ACT score of students was directly compared with their final letter grade from the student’s introductory physics class. The student …

Simulating And Modelling Opinion Dynamics, Jennifer Heermance

Graduate Research Theses & Dissertations

The foundation of social media is conversation. Social media allows people to share ideas and opinions, as well as discuss those opinions. A point of intrigue for many social scientists is how those opinions change through interaction with others. What influences someone’s opinion? When is a person willing to adapt their opinion, and when does it remain the same? Is it possible to measure these opinion dynamics? Our overall goal is to develop a more comprehensive model for opinion dynamics. The first step of this process is to simulate data that can then be analyzed and used to develop a …

Using Reinforcement Learning In A Simulated Intelligent Tutoring System, Manohar Sai Jasti

Graduate Research Theses & Dissertations

I used reinforcement learning to investigate which categories of hints are most efficient in an intelligent tutoring system for human anatomy. Efficiency is defined as minimizing the time it takes the student to learn the material. When a student gives a wrong answer, the tutor can give them a text hint, a diagrammatic hint, or a video clip. Each type of hint takes a different amount of time to deliver and takes the student a different amount of time to understand.

I built a simulator for the intelligent tutoring system to collect data from simulated students. I implemented reinforcement learning, …

A Survey Of Big Data Tools And Technologies, Mrinal Kanti Roy

Graduate Research Theses & Dissertations

Big data analytics require the use of different sets of tools and technologies. This thesis categorizes big data analytics in three areas: storage, processing, and visualization and provides discussions of available tools and technologies for each area. It also provides a hands-on review of available tools and review of relevant research papers for such tools and technologies. The goal of this thesis is to summarize and organize current big data tools and technologies by surveying recent research papers, provide a synthesis and discuss observation current issues and future trends.

Water Quality And Small Unmanned Aerial System Photogrammetry Of Boeng Cheung Ek Wastewater Treatment Wetland, Phnom Penh, Cambodia, Samuel Vesal Mallow

Graduate Research Theses & Dissertations

Boeng Cheung Ek wetland is a large, peri-urban, free water surface, natural wastewater treatment wetland that services approximately 90% of combined storm water and untreated sewage from the city of Phnom Penh, Cambodia, population over 1.5 million. Due to monsoon rainfall and connectivity to the Mekong River, the wetted area cycles between 13 km2 and 20 km2 annually. The objective of this study is to characterize the spatiotemporal variability of contaminants in wetland surface water and groundwater. Dissolved ions including ammonium, nitrate-N (as NO3), phosphate, and chloride, along with E. coli, trace metals, δH, δ18O, and physicochemical parameters were determined …

Tools For Understanding Static Structure Factors And Their Application To Simulations Of Liquids, Travis Mackoy

Graduate Research Theses & Dissertations

Molecular dynamics (MD) simulations can be used to compute static structure factors (��(��)) and provide an interpretation of the underlying periodic atomic ordering. MD simulations complement experimentally measured ��(��) by allowing qualitative assignment of peaks to various ordering, such as cation-anion ordering in ionic liquids, via decomposition of ��(��) into partial ��(��). Here we present a method for classifying interatomic distances that allows for quantitative peak assignment and visualization of atoms that contribute most to each peak in calculated ��(��) for soft materials. The method is illustrated by investigating ��(��) for the ionic liquid 1-butyl-1-methyl-pyrrolidinium bis(trifluoromethylsulfonyl)imide (C4C1pyrrTFSI), which shows two …

Study Of Electron Beam Emitted From Nano-Structured Cathode, Anusorn Lueangaramwong

Graduate Research Theses & Dissertations

Over the years, particle accelerators have enabled breakthroughs in fundamental science while also becoming critical tools in medicine, homeland security, and societal applications. High-brightness electron beams are a key to the development of some of these accelerators and especially accelerator-based light sources. Electron emission is the first stage of the accelerator that dictates the ultimate beam brightness. To achieve such beam quality, electron sources--cathodes have been developed for their emission efficiencies. The cathodes with nanostructures on their surfaces have shown promising efficiencies when a laser pulse impinges, or an electric field is applied, on the cathode surface. Alternatively, these nano-patterned …

Towards Dual-Readout Calorimetry For Redtop Experiment, Tilak B. Malla

Graduate Research Theses & Dissertations

By virtue of peculiar properties of the eta (η) Goldstone boson, the study of it’s de-

cays is very important. The rare decays of this meson can provide unique opportunities

to probe several fundamental invariance principles of physics world. The REDTOP experi-

ment aims at searching for Physics Beyond Standard Model (BSM) by studying such decays

at a sensitivity of 10^ (−10) or below, which requires 10^ (12) or more η mesons. The REDTOP

collaboration has designed a detector to produce and reconstruct such a large number of

η mesons, which will make REDTOP a rare decay factory. The detector …

Assessing The Performance And Merit Of The Random Survival Forest And Cox Models On A Pancreatic Cancer Data Set, Carl Edward Mueller

Graduate Research Theses & Dissertations

Random Survival Forest (RSF) is one of the most powerful and easily applied machine learning models for survival data. RSF sacrifices some of the interpretability of the decision trees used to grow the forest in order to significantly reduce the bias and variance of the basic classification and regression tree (CART) paradigm. The lessened interpretability and higher computational intensity of RSF means that it may not always be the preferred method, even in settings where black-box methods are readily used. By contrast, the Cox Proportional Hazards (PH) model is incredibly flexible, resistant to overfitting, and transparently estimable. The tradeoff for …

Bayesian Lasso Survival Analysis, Justin P. Neely

Graduate Research Theses & Dissertations

This thesis examines the use of Bayesian LASSO regression for survival data to estimate the survival function and to select significant covariates simultaneously. We consider survival times of patients with adenocarcinoma lung cancer. The survival and genetic data are available in the Cancer Genome Atlas (TCGA) Research Network. As a pilot study, within chromosome 5, we apply Bayesian LASSO regression to explore genetic markers that may help to identify crucial genes to determine survival times of patients. Using Gibbs sampling we can obtain Markov Chain Monte Carlo samples for regression coefficients and model variance as well as LASSO penalty from …

Microbiome And Metagenome Of An Anthropogenic Hyperalkaline Slag-Fill Waterway In South Chicago, Illinois, Usa, Jan Ingemar Ohlsson

Graduate Research Theses & Dissertations

The aquifers and waterways around the Lake Calumet area of south Chicago, IL, USA have been affected by a century of intensive industrial waste dumping. Steel slag in particular was used as infilling to claim most of the Calumet Wetlands for industrial and residential development. Since the prohibition and cessation of these disposal activities, a growing body of research has recorded the environmental and health effects of the infill, particularly where interactions between slag minerals and rainwater produces extremely alkaline groundwater (pH ≥13). Atmospheric carbon dioxide interacting with these calcium hydroxide-rich slag fluids produces massive precipitation of a loose, white …

Experimental And Computational Studies Of Gas-Phase Metal Ion Catalysts, Elettra Ludovica Piacentino

Graduate Research Theses & Dissertations

In this thesis, the chemistry of three groups of transition metal coordination complexes

is reported: A series of zinc–based catalysts was evaluated for efficiency in decomposing

formic acid into molecular hydrogen and carbon dioxide in the gas phase. The effectiveness

of the catalysts in the series [LZn(H)]+, where L = terpyridine (tpy), phenantroline (phen)

or 2,2’-bipyrydine (bpy) was found to depend on the ligand used, which turned out to be fun

damental in tuning the catalytic properties of the zinc complex. Specifically, [(tpy)Zn(H)]+

displayed the fastest formic acid decomposition as observed inside a quadrupole ion trap mass

spectrometer. The catalysts …

Coupling Reactions Of Triaryl Carbocations Via Cationic Charge Delocalization, Hongfang Xue

Graduate Research Theses & Dissertations

This thesis describes a coupling reaction involving cationic charge delocalization. This work began as a study of the Piancatelli rearrangement, a reaction involving furan-based triaryl methyl cations. A new synthetic reaction has been discovered using these furan-based triaryl methyl cations as electrophilic intermediates. Zinc chloride has been found to be an effective Lewis acid catalyst for the ionization of the triaryl methanols to the carbocations. It has been found that furan-based triaryl methyl cations react with nucleophiles such as enol silyl ethers and silyl ketene acetals. Similar to the Piancatelli reaction, nucleophilic reactions are found to occur at the 5-position …

Visualization Of Large Time-Dependent Multidimensional Datasets Using Information Dashboards, Ankita Upadhyay

Graduate Research Theses & Dissertations

Visual summarization is the act of displaying the most important information in a single

view or on a single screen. There are many existing powerful and useful information visualization tools and techniques to visualize large datasets, but the major challenge in using

these visualization tools efficiently are assessing what is needed to create awareness,

where awareness is dependent on the situation and context of the user and surrounding

events.

For example, when a user is driving (context) and approaching an intersection (situation)

the traffic light being red creates an awareness that the driver needs to stop, or when a fire …

Direct Polymer Grafting As A Method Of Maintaining The Mechanical Properties Of Cellulose Nanocrystals In The Presence Of Moisture, Mary Elizabeth Breen-Lyles

Graduate Research Theses & Dissertations

Cellulose nanocrystals (CNCs) are a distinctive nanomaterial derived from cellulose, the most abundant natural polymer on Earth, and the primary reinforcing structural component of cellulose fibrils found within the plant cell wall. These nanocrystals exhibit mechanical properties comparable to synthetic aramid fibers but are advantageous as they are biodegradable, renewable, and can be produced sustainably as they are predominantly extracted from naturally occurring cellulosic materials. These qualities make it a sustainable, highly renewable and environmentally friendly material to be used in place of synthetic materials in a variety of applications. With their high surface area to volume ratio, low level …

Development Of Chromatography And Mass Spectrometry Methods For Enzyme Kinetics And Inhibition In Glycolysis And The Methylerythritol Phosphate (Mep) Pathways, Allison Carr

Graduate Research Theses & Dissertations

This dissertation focuses on method development for enzymatic activity and inhibition assays for use in screening potential antibacterial compounds. Infection by antibiotic-resistant pathogens in on the rise globally and several enzymes, including the those in the methylerythritol phosphate (MEP) pathway and fructose bisphosphate aldolase, as strong targets for novel pharmaceuticals because they are not present in humans. Several assays exist for the MEP pathway and aldolase, but there is always work to be done to improve sensitivity, increase throughput, and lower cost. This dissertation describes the development of several methods – hydrophilic interaction chromatography (HILIC), liquid chromatography-mass spectrometry (LC-MS), and …

Bivariate Cure Rate Model Using Copula Functions In Presence Of Censored Data And Covariates, Jie Huang

Graduate Research Theses & Dissertations

Bivariate survival cure rate models extend the understanding of time-to-event data by allowing for the formulation of more accurate and informative conclusion. These conclusions are obtainable from an analysis that accounts for a cured fraction of the population and dependence between paired units. We propose a mixture cure rate model where a correlation coefficient is used for the association between bivariate cure rate fractions and a new generalized Farlie Gumbel Morgenstern (FGM) copula function is applied to model the de-

pendence structure of bivariate survival times. Covariate effects are incorporated into two components of our model, cure rate fractions and …

Applications Of Bayesian Functional Data Analysis, Hao Shen

Graduate Research Theses & Dissertations

Functional Data Analysis (FDA) is a set of statistical methods that can deal with the data which represent curves or functions. In this dissertation, we consider two extensions of FDA to two types of data, circadian data and multidimensional data. The first part of the dissertation is concerned with the analysis of circadian data. We estimate circadian functions by using Bayesian smoothing splines under the generalized linear model, and extract two measures from each estimated function, magnitude and roughness. Based on extracted measures, we cluster individual functions into normal group and abnormal group by utilizing a density based clustering method. …