Statistics and Probability Commons^™

Integrating Machine Learning Methods For Medical Diagnosis, Jazmin Quezada

Open Access Theses & Dissertations

Abstract:The rapid advancement of machine learning techniques has revolutionized the field of medical diagnosis by offering powerful tools to analyze complex data sets and make accurate predictions. In this proposed method, we present a novel approach that integrates machine learning and optimization models to enhance the accuracy of medical diagnoses. Our method focuses on fine-tuning and optimizing the parameters of machine learning algorithms commonly used in medical diagnosis, such as logistic regression, support vector machines, and neural networks. By employing optimization techniques, we systematically explore the parameter space of these algorithms to discover the most optimal configurations. Moreover, by representing …

Generative Adversarial Game With Tailored Quantum Feature Maps For Enhanced Classification, Anais Sandra Nguemto Guiawa

Doctoral Dissertations

In the burgeoning field of quantum machine learning, the fusion of quantum computing and machine learning methodologies has sparked immense interest, particularly with the emergence of noisy intermediate-scale quantum (NISQ) devices. These devices hold the promise of achieving quantum advantage, but they grapple with limitations like constrained qubit counts, limited connectivity, operational noise, and a restricted set of operations. These challenges necessitate a strategic and deliberate approach to crafting effective quantum machine learning algorithms.

This dissertation revolves around an exploration of these challenges, presenting innovative strategies that tailor quantum algorithms and processes to seamlessly integrate with commercial quantum platforms. A …

Traditional Vs Machine Learning Approaches: A Comparison Of Time Series Modeling Methods, Miguel E. Bonilla Jr., Jason Mcdonald, Tamas Toth, Bivin Sadler

SMU Data Science Review

In recent years, various new Machine Learning and Deep Learning algorithms have been introduced, claiming to offer better performance than traditional statistical approaches when forecasting time series. Studies seeking evidence to support the usage of ML/DL over statistical approaches have been limited to comparing the forecasting performance of univariate, linear time series data. This research compares the performance of traditional statistical-based and ML/DL methods for forecasting multivariate and nonlinear time series.

Stressor: An R Package For Benchmarking Machine Learning Models, Samuel A. Haycock

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Many discipline specific researchers need a way to quickly compare the accuracy of their predictive models to other alternatives. However, many of these researchers are not experienced with multiple programming languages. Python has recently been the leader in machine learning functionality, which includes the PyCaret library that allows users to develop high-performing machine learning models with only a few lines of code. The goal of the stressor package is to help users of the R programming language access the advantages of PyCaret without having to learn Python. This allows the user to leverage R’s powerful data analysis workflows, while simultaneously …

An Interval-Valued Random Forests, Paul Gaona Partida

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

There is a growing demand for the development of new statistical models and the refinement of established methods to accommodate different data structures. This need arises from the recognition that traditional statistics often assume the value of each observation to be precise, which may not hold true in many real-world scenarios. Factors such as the collection process and technological advancements can introduce imprecision and uncertainty into the data.

For example, consider data collected over a long period of time, where newer measurement tools may offer greater accuracy and provide more information than previous methods. In such cases, it becomes crucial …

A Machine Learning Approach For Predicting Clinical Trial Patient Enrollment In Drug Development Portfolio Demand Planning, Ahmed Shoieb

Masters Theses

One of the biggest challenges the clinical research industry currently faces is the accurate forecasting of patient enrollment (namely if and when a clinical trial will achieve full enrollment), as the stochastic behavior of enrollment can significantly contribute to delays in the development of new drugs, increases in duration and costs of clinical trials, and the over- or under- estimation of clinical supply. This study proposes a Machine Learning model using a Fully Convolutional Network (FCN) that is trained on a dataset of 100,000 patient enrollment data points including patient age, patient gender, patient disease, investigational product, study phase, blinded …

Examining Model Complexity's Effects When Predicting Continuous Measures From Ordinal Labels, Mckade S. Thomas

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Many real world problems require the prediction of ordinal variables where the values are a set of categories with an ordering to them. However, in many of these cases the categorical nature of the ordinal data is not a desirable outcome. As such, regression models treat ordinal variables as continuous and do not bind their predictions to discrete categories. Prior research has found that these models are capable of learning useful information between the discrete levels of the ordinal labels they are trained on, but complex models may learn ordinal labels too closely, missing the information between levels. In this …

Distance Correlation Based Feature Selection In Random Forest, Jose Munoz-Lopez

Electronic Theses, Projects, and Dissertations

The Pearson correlation coefficient is a commonly used measure of correlation, but it has limitations as it only measures the linear relationship between two numerical variables. In 2007, Szekely et al. introduced the distance correlation, which measures all types of dependencies between random vectors X and Y in arbitrary dimensions, not just the linear ones. In this thesis, we propose a filter method that utilizes distance correlation as a criterion for feature selection in Random Forest regression. We conduct extensive simulation studies to evaluate its performance compared to existing methods under various data settings, in terms of the prediction mean …

Application Of Sentiment Analysis And Machine Learning Techniques To Predict Daily Cryptocurrency Price Returns, Edward Wu

CMC Senior Theses

This paper examines the effects of social media sentiment relating to Bitcoin on the daily price returns of Bitcoin and other popular cryptocurrencies by utilizing sentiment analysis and machine learning techniques to predict daily price returns. Many investors think that social media sentiment affects cryptocurrency prices. However, the results of this paper find that social media sentiment relating to Bitcoin does not add significant predictive value to forecasting daily price returns for each of the six cryptocurrencies used for analysis and that machine learning models that do not assume linearity between the current day price return and previous daily price …

Application Of Distributed Fiber-Optic Sensing For Pressure Predictions And Multiphase Flow Characterization, Gerald Kelechi Ekechukwu

LSU Doctoral Dissertations

In the oil and gas industry, distributed fiber optics sensing (DFOS) has the potential to revolutionize well and reservoir surveillance applications. Using fiber optic sensors is becoming increasingly common because of its chemically passive and non-magnetic interference properties, the possibility of flexible installations that could be behind the casing, on the tubing, or run on wireline, as well as the potential for densely distributed measurements along the entire length of the fiber. The main objectives of my research are to develop and demonstrate novel signal processing and machine learning computational techniques and workflows on DFOS data for a variety of …

Machine Learning To Predict Warhead Fragmentation In-Flight Behavior From Static Data, Katharine Larsen

Doctoral Dissertations and Master's Theses

Accurate characterization of fragment fly-out properties from high-speed warhead detonations is essential for estimation of collateral damage and lethality for a given weapon. Real warhead dynamic detonation tests are rare, costly, and often unrealizable with current technology, leaving fragmentation experiments limited to static arena tests and numerical simulations. Stereoscopic imaging techniques can now provide static arena tests with time-dependent tracks of individual fragments, each with characteristics such as fragment IDs and their respective position vector. Simulation methods can account for the dynamic case but can exclude relevant dynamics experienced in real-life warhead detonations. This research leverages machine learning methodologies to …

Classification Of Pixel Tracks To Improve Track Reconstruction From Proton-Proton Collisions, Kebur Fantahun, Jobin Joseph, Halle Purdom, Nibhrat Lohia

SMU Data Science Review

In this paper, machine learning techniques are used to reconstruct particle collision pathways. CERN (Conseil européen pour la recherche nucléaire) uses a massive underground particle collider, called the Large Hadron Collider or LHC, to produce particle collisions at extremely high speeds. There are several layers of detectors in the collider that track the pathways of particles as they collide. The data produced from collisions contains an extraneous amount of background noise, i.e., decays from known particle collisions produce fake signal. Particularly, in the first layer of the detector, the pixel tracker, there is an overwhelming amount of background noise that …

Mathematical Models Yield Insights Into Cnns: Applications In Natural Image Restoration And Population Genetics, Ryan Cecil

Electronic Theses and Dissertations

Due to a rise in computational power, machine learning (ML) methods have become the state-of-the-art in a variety of fields. Known to be black-box approaches, however, these methods are oftentimes not well understood. In this work, we utilize our understanding of model-based approaches to derive insights into Convolutional Neural Networks (CNNs). In the field of Natural Image Restoration, we focus on the image denoising problem. Recent work have demonstrated the potential of mathematically motivated CNN architectures that learn both `geometric' and nonlinear higher order features and corresponding regularizers. We extend this work by showing that not only can geometric features …

Better Understanding Genomic Architecture With The Use Of Applied Statistics And Explainable Artificial Intelligence, Jonathon C. Romero

Doctoral Dissertations

With the continuous improvements in biological data collection, new techniques are needed to better understand the complex relationships in genomic and other biological data sets. Explainable Artificial Intelligence (X-AI) techniques like Iterative Random Forest (iRF) excel at finding interactions within data, such as genomic epistasis. Here, the introduction of new methods to mine for these complex interactions is shown in a variety of scenarios. The application of iRF as a method for Genomic Wide Epistasis Studies shows that the method is robust in finding interacting sets of features in synthetic data, without requiring the exponentially increasing computation time of many …

Attempting To Predict The Unpredictable: March Madness, Coleton Kanzmeier

Theses/Capstones/Creative Projects

Each year, millions upon millions of individuals fill out at least one if not hundreds of March Madness brackets. People test their luck every year, whether for fun, with friends or family, or to even win some money. Some people rely on their basketball knowledge whereas others know it is called March Madness for a reason and take a shot in the dark. Others have even tried using statistics to give them an edge. I intend to follow a similar approach, using statistics to my advantage. The end goal is to predict this year’s, 2022, March Madness bracket. To achieve …

Searching For Anomalous Extensive Air Showers Using The Pierre Auger Observatory Fluorescence Detector, Andrew Puyleart

Dissertations, Master's Theses and Master's Reports

Anomalous extensive air showers have yet to be detected by cosmic ray observatories. Fluorescence detectors provide a way to view the air showers created by cosmic rays with primary energies reaching up to hundreds of EeV . The resulting air showers produced by these highly energetic collisions can contain features that deviate from average air showers. Detection of these anomalous events may provide information into unknown regions of particle physics, and place constraints on cross-sectional interaction lengths of protons. In this dissertation, I propose measurements of extensive air shower profiles that are used in a machine learning pipeline to distinguish …

Forecasting Bitcoin, Ethereum And Litecoin Prices Using Machine Learning, Sai Prabhu Jaligama

Graduate Research Theses & Dissertations

This research aims to predict the cryptocurrencies Bitcoin, Litecoin and Ethereum using Time Series Modelling with daily data of closing price from 16th of October 2018 to 9th of September 2021for a total of 1073 days. Augmented Dickey Fuller test was first used to check stationarity of the time series, then two forecasting algorithms called ARIMA, and PROPHET were used to make predictions. The findings show similar results for both the models for each of Bitcoin, Ethereum and Litecoin. The results achieved show modelling cryptocurrencies which are volatile using a single variable produces satisfying results.

Reinforcement Learning: Low Discrepancy Action Selection For Continuous States And Actions, Jedidiah Lindborg

Electronic Theses and Dissertations

In reinforcement learning the process of selecting an action during the exploration or exploitation stage is difficult to optimize. The purpose of this thesis is to create an action selection process for an agent by employing a low discrepancy action selection (LDAS) method. This should allow the agent to quickly determine the utility of its actions by prioritizing actions that are dissimilar to ones that it has already picked. In this way the learning process should be faster for the agent and result in more optimal policies.

Statistical Modeling, Learning And Computing For Stochastic Dynamics Of Complex Systems, Mohammadmahdi Hajiha

Graduate Theses and Dissertations

With the recent advances in sensor technology, it is much easier to collect and store streams of system operational and environmental (SOE) data. These data can be used as input to model the underlying behavior of complex engineered systems and phenomenons if appropriate algorithms with well-defined assumptions are developed. This dissertation is comprised of the research work to show the applicability of SOE data when fed into proposed tailored algorithms. The first purposes of these algorithms are to estimate and analyze the reliability of a system as elaborated in Chapter 2. This chapter provides the derivation of closed-form expressions that …

Online And Adjusted Human Activities Recognition With Statistical Learning, Yanjia Zhang

USF Tampa Graduate Theses and Dissertations

Wearable human activity recognition (HAR) is a widely application system for our daily life. It hasbeen built in many devices, such as smartphone, smartwatch, activity tracker, and health monitor. Many researchers try to develop a system which requires less memory space and power, but has fast and accurate classification results. Moreover, the objective of adjusting the classifier by the system self is also a study direction. In the present study, we introduced the machine learning methods to both smartphone data and smartwatch data and an adjusted model with the continuous generating data. Further, we also proposed a new HAR system …

Machine Learning Based Restaurant Sales Forecasting, Austin B. Schmidt

University of New Orleans Theses and Dissertations

To encourage proper employee scheduling for managing crew load, restaurants have a need for accurate sales forecasting. We predict partitions of sales days, so each day is broken up into three sales periods: 10:00 AM-1:59 PM, 2:00 PM-5:59 PM, and 6:00 PM-10:00 PM. This study focuses on the middle timeslot, where sales forecasts should extend for one week. We gather three years of sales between 2016-2019 from a local restaurant, to generate a new dataset for researching sales forecasting methods.

Outlined are methodologies used when going from raw data to a workable dataset. We test many machine learning models on …

Node Classification On Relational Graphs Using Deep-Rgcns, Nagasai Chandra

Master's Theses

Knowledge Graphs are fascinating concepts in machine learning as they can hold usefully structured information in the form of entities and their relations. Despite the valuable applications of such graphs, most knowledge bases remain incomplete. This missing information harms downstream applications such as information retrieval and opens a window for research in statistical relational learning tasks such as node classification and link prediction. This work proposes a deep learning framework based on existing relational convolutional (R-GCN) layers to learn on highly multi-relational data characteristic of realistic knowledge graphs for node property classification tasks. We propose a deep and improved variant, …

Machine Learning Morphisms: A Framework For Designing And Analyzing Machine Learning Work Ows, Applied To Separability, Error Bounds, And 30-Day Hospital Readmissions, Eric Zenon Cawi

McKelvey School of Engineering Theses & Dissertations

A machine learning workflow is the sequence of tasks necessary to implement a machine learning application, including data collection, preprocessing, feature engineering, exploratory analysis, and model training/selection. In this dissertation we propose the Machine Learning Morphism (MLM) as a mathematical framework to describe the tasks in a workflow. The MLM is a tuple consisting of: Input Space, Output Space, Learning Morphism, Parameter Prior, Empirical Risk Function. This contains the information necessary to learn the parameters of the learning morphism, which represents a workflow task. In chapter 1, we give a short review of typical tasks present in a workflow, as …

Feature Investigation For Stock Returns Prediction Using Xgboost And Deep Learning Sentiment Classification, Seungho (Samuel) Lee

CMC Senior Theses

This paper attempts to quantify predictive power of social media sentiment and financial data in stock prediction by utilizing a comprehensive set of stock-related fundamental and technical variables and social media sentiments. For conducting sentiment analysis, this study employs a pretrained finBERT model that provides three different sentiment classifications and respective softmax scores. Hence, the significance of these variables is evaluated with XGBoost regression and Shapley Additive exPlanations (SHAP) frameworks. Through investigating feature importance, this study finds that statistical properties of sentiment variables provide a stronger predictive power than a weighted sentiment score and that it is possible to quantify …

Statistical And Machine Learning Approaches To Depressive Disorders Among Adults In The United States: From Factor Discovery To Prediction Evaluation, Minhwa Lee

Senior Independent Study Theses

According to the National Institutes of Mental Health (NIMH), depressive disorders (or major depression) are considered one of the most common and serious health risks in the United States. Our study focuses on extracting non-medical factors of depressive disorders diagnosis, such as overall health states, health risk behaviors, demography, and healthcare access, using the Behavioral Risk Factor Surveillance System (BRFSS) data set collected by the Centers for Disease Control and Prevention (CDC) in 2018.

We set the two objectives of our study about depressive disorders diagnosis in the United States as follows. First, we aim to utilize machine learning algorithms …

Random Search Plus: A More Effective Random Search For Machine Learning Hyperparameters Optimization, Bohan Li

Masters Theses

Machine learning hyperparameter optimization has always been the key to improve model performance. There are many methods of hyperparameter optimization. The popular methods include grid search, random search, manual search, Bayesian optimization, population-based optimization, etc. Random search occupies less computations than the grid search, but at the same time there is a penalty for accuracy. However, this paper proposes a more effective random search method based on the traditional random search and hyperparameter space separation. This method is named random search plus. This thesis empirically proves that random search plus is more effective than random search. There are some case …

Machine Learning Applications For Drug Repurposing, Hansaim Lim

Dissertations, Theses, and Capstone Projects

The cost of bringing a drug to market is astounding and the failure rate is intimidating. Drug discovery has been of limited success under the conventional reductionist model of one-drug-one-gene-one-disease paradigm, where a single disease-associated gene is identified and a molecular binder to the specific target is subsequently designed. Under the simplistic paradigm of drug discovery, a drug molecule is assumed to interact only with the intended on-target. However, small molecular drugs often interact with multiple targets, and those off-target interactions are not considered under the conventional paradigm. As a result, drug-induced side effects and adverse reactions are often neglected …

Compressed Dna Representation For Efficient Amr Classification, John Partee, Robert Hazell, Anjli Solsi, John Santerre

SMU Data Science Review

In this paper, we explore a representation methodology for the compression of DNA isolates. Using lossless string compression via tokenization of frequently repeated segments of DNA, we reduce the length of the isolates to be counted as k-mers for classification. With this new representation, we apply a previously established feature sampling method to dramatically reduce the feature space. In understanding the genetic diversity, we also look at conserving biological function across these spaces. Using a random forest model we were able to predict the resistance or susceptibility of bacteria with 85-90\% accuracy, with a 30-50\% reduction in overall isolate length, …

Evaluation Of Standard And Semantically-Augmented Distance Metrics For Neurology Patients, Daniel B. Hier, Jonathan Kopel, Steven U. Brint, Donald C. Wunsch, Gayla R. Olbricht, Sima Azizi, Blaine Allen

Electrical and Computer Engineering Faculty Research & Creative Works

Background: Patient distances can be calculated based on signs and symptoms derived from an ontological hierarchy. There is controversy as to whether patient distance metrics that consider the semantic similarity between concepts can outperform standard patient distance metrics that are agnostic to concept similarity. The choice of distance metric can dominate the performance of classification or clustering algorithms. Our objective was to determine if semantically augmented distance metrics would outperform standard metrics on machine learning tasks.

Methods: We converted the neurological findings from 382 published neurology cases into sets of concepts with corresponding machine-readable codes. We calculated patient distances by …

Machine-Learning-Based Prediction Of Sepsis Events From Vertical Clinical Trial Data: A Naïve Approach, Tyler Michael Gaddis

Theses and Dissertations

Sepsis is a potentially life-threatening condition characterized by a dysregulated, disproportionate immune response to infection by which the afflicted body attacks its own tissues, sometimes to the point of organ failure, and in the worst cases, death. According to the Centers for Disease Control and Prevention (CDC) Sepsis is reported to kill upwards of 270,000 Americans annually, though this figure may be greater given certain ambiguities in the current accepted diagnostic framework of the disease.

This study attempted to first establish an understanding of past definitions of sepsis, and to then recommend use of machine learning as integral in an …