Physical Sciences and Mathematics Commons^™

Using Social Network Analysis To Measure And Visualize Student Clustering Within Middle And High Schools, Geoffrey David West

USF Tampa Graduate Theses and Dissertations

The dominant philosophy of American public schools has been to group students together based on similar characteristics. Known as tracking, high achieving students would take courses on the “college track” while others would take “career track” courses. It was not long until advocates noticed that this process unfairly advantaged affluent and White student over poor and minoritized groups. A new process called “ability grouping” took over where tracking left off, but to the same effect. It is difficult to measure the degree students are grouped together by a certain characteristic, and while a few research papers aim to do so, …

Exploring Time-Varying Extraneous Variables Effects In Single-Case Studies, Ke Cheng

USF Tampa Graduate Theses and Dissertations

The effect of time-varying extraneous variables has been studied in other statistical analyses such as using Kaplan–Meier or Cox regression analysis in survival analyses. Nonetheless, the effect of modeling versus not modeling individual specific time varying extraneous variables has not been explored in multiple-baseline single case designs through Monte Carlo simulation studies. Therefore, in my dissertation, I used simulation methods to explore for a variety of conditions (varying in the number of participants, number of observations per participant, type of extraneous variable effect, size of the true intervention effect) the impact of extraneous variables on bias and standard error of …

Fuzzy Kc Clustering Imputation For Missing Not At Random Data, Markku A. Malmi Jr.

USF Tampa Graduate Theses and Dissertations

Research has a variety of difficulties, especially when involving human subjects, and one of the most prevalent is the issue of missing data. Missing data will always be present in research due to the fact there is no perfect method for collecting data and protecting against human error or mechanical failure. This requires researchers to be able to mitigate the problems that come along with missing data; reduction in power of an analysis and bias introduced by the missing pattern. This research investigated a non-parametric method using a nested approach of fuzzy K-Modes and fuzzy C-Means clustering to impute missing …

Association Between Use Of Remdesivir And Bradycardia, Gibret Umeukeje

USF Tampa Graduate Theses and Dissertations

Remdesivir received the first emergency use authorization from the FDA for the treatment of COVID-19. Multiple adverse drug reactions (ADR) have been reported since its approval in October 2020. Bradycardia, defined by a decrease in heart rate has been reported as an adverse event for patients receiving remdesivir for COVID-19 treatment. The purpose of the research is to systematically investigate the frequency of occurrence of bradycardia in adults receiving remdesivir using clinical data derived from the FDA Adverse Event Reporting System (FAERS) database. Patients receiving remdesivir were compared to those receiving Paxlovid, Regen-Cov, and Dexamethasone for COVID-19 treatment to see …

Data-Driven Analytical Predictive Modeling For Pancreatic Cancer, Financial & Social Systems, Aditya Chakraborty

USF Tampa Graduate Theses and Dissertations

Pancreatic cancer is one of the most deathly disease and becoming an increasingly commoncause of cancer mortality. It continues giving rise to massive challenges to clinicians and cancer researchers. The combined five-year survival rate for pancreatic cancer is extremely low, about 5 to 10 percent, owing to the fact that a large number of the patients are diagnosed at stage IV when the disease has metastasized. Our study investigates if there exists any statistical significant difference between the median survival times and also the survival probabilities of male and female pancreatic cancer patients at different cancer stages, and irrespective of …

Nonparametric Estimation Of Transition Probabilities In Illness-Death Model Based On Ranked Set Sampling, Ying Ma

USF Tampa Graduate Theses and Dissertations

The ranked set sampling (RSS) design is applied widely in agriculture, environmental science, and medical research where the exact measurements of sampling units is costly, but sampling units can be ranked by a correlated concomitant variable. RSS is usually a cost-efficient alternate to simple random sampling (SRS) for selecting more representative samples. This study presents a novel methodology to investigate the nonparametric estimation of transition probabilities in illness-death model using the RSS design. We study the Aalen–Johansen estimator of transition probabilities in illness-death Markov model based on RSS design under random right censoring time and propose nonparametric estimators of the …

New Developments In Statistical Optimal Designs For Physical And Computer Experiments, Damola M. Akinlana

USF Tampa Graduate Theses and Dissertations

Statistical design of experiments allows for multiple factors influencing a process to be systematically manipulated in an experiment, and their effects on the output of the process to be studied via statistical modeling and analysis. Classical designs offer general nice performance but have limited applications due to restricted design size, region, and randomization structure. Computer generated optimal designs become more popular in recent decades due to the rapid growth in computing power. Most existing work in optimal design of experiments involves designing experiments with optimal performance on a single chosen objective or a single response. However, with the increasing limitation …

Video Anomaly Detection: Practical Challenges For Learning Algorithms, Keval Doshi

USF Tampa Graduate Theses and Dissertations

Anomaly detection in surveillance videos is attracting an increasing amount of attention. Despite the competitive performance of several existing methods, they lack theoretical performance analysis, particularly due to the complex deep neural network architectures used in decision making. Additionally, real-time decision making is an important but mostly neglected factor in this domain. Much of the existing methods that claim to be online, depend on batch or offline processing in practice. Furthermore, several critical tasks such as continual learning, model interpretability and cross-domain adaptability are completely neglected in existing works. Motivated by these research gaps, in this dissertation we discuss our …

A Functional Optimization Approach To Stochastic Process Sampling, Ryan Matthew Thurman

USF Tampa Graduate Theses and Dissertations

The goal of the current research project is the formulation of a method for the estimation and modeling of additive stochastic processes with both linear- and cycle-type trend components as well as a relatively robust noise component in the form of Levy processes. Most of the research in stochastic processes tends to focus on cases where the process is stationary, a condition that cannot be assumed for the model above due to the presence of the cyclical sub-component in the overall additive process. As such, we outline a number of relevant theoretical and applied topics, such as stochastic processes and …

Using Fine-Scale Aquatic Habitat Data To Construct Dreissenid Sdms In The Laurentian Great Lakes, Grace C. Henderson

USF Tampa Graduate Theses and Dissertations

The invasion of the Laurentian Great Lakes by aquatic invasive species (AIS) has been the subject of investigation for decades, due to their dramatic alterations to the ecosystem and high economic costs. Two AIS with the largest impacts are dreissenid zebra and quagga mussels, and though these species have been studied extensively, questions remain about what factors control their distributions, and whether lake warming will alter these distributions. Species distribution models (SDMs) offer a powerful tool to examine the relationship between species presences and environmental variables, which are typically bioclimactic data. The creation of the Aquatic Habitat (AqHab) dataset containing …

Measurements Of Generalizability And Adjustment For Bias In Clinical Trials, Yuanyuan Lu

USF Tampa Graduate Theses and Dissertations

While randomized controlled trials (RCTs) are widely used as a gold standard in clinical research and public health, they are criticized because of a potential lack of generalizability, as the trial patients may be unrepresentative of the target patient population. Few research addresses how to assess and evaluate the generalizability of RCTs. As we know, patients are rarely selected on a random basis from a well-defined patient population of interest into a clinical trial. Generalizing findings from the RCT samples to the patient population has begun to receive increasing attention. We simulate a patient population with treatment effect size of …

Uncertainty Quantification In Deep And Statistical Learning With Applications In Bio-Medical Image Analysis, K. Ruwani M. Fernando

USF Tampa Graduate Theses and Dissertations

Deep Learning (DL) has achieved the state-of-the-art performance across a broad spectrum oftasks. From a statistical standpoint, deep neural networks can be construed as universal function approximators. Although statistical modeling and deep learning methods are well-established as independent areas of research, hybridization of the two paradigms via probabilistic deep networks is an emerging trend. Through development of novel analytical methods under the statistical and deep-learning framework, we address some of the major challenges encountered in the design of intelligent systems which include class imbalance learning, probability calibration, uncertainty quantification and high dimensionality. When modeling rare events, existing methodologies require re-sampling …

Differential Privacy For Regression Modeling In Health: An Evaluation Of Algorithms, Joseph Ficek

USF Tampa Graduate Theses and Dissertations

Background: There is a need for rigorous and standardized methods of privacy protection for shared data in the health sciences. Differential privacy is one such method that has gained much popularity due to its versatility and robustness. This study evaluates differential privacy for explanatory regression modeling in the context of health research.

Methods: Surveyed and newly proposed algorithms were evaluated with respect to the accuracy (bias and RMSE) of coefficient estimates, the empirical coverage probability of confidence intervals, and the power and type I error rates of hypothesis tests. Evaluations took place in both simulated and real data from a …

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 …

Development And Validation Of A Scale To Measure Songwriting Self-Efficacy (Sses) With Secondary Music Students, Patrick K. Cooper

USF Tampa Graduate Theses and Dissertations

Social cognitive theory was developed to explain how individuals learn, in part, by witnessing the behavior of others. Self-efficacy is a construct within social cognitive theory which indicates the beliefs that an individual can be successful at a task under specific situational demands. The sources of self-efficacy include self-evaluating past experiences to predict future success, comparing our abilities to those around us, the verbal and social feedback we get from others, and the physiological feelings we experience when engaged in or thinking about the task. Measures of self-efficacy have been shown to be accurate predictors of successful learning outcomes, achievement, …

Bayesian Multivariate Joint Modeling For Skewed-Longitudinal And Time-To-Event Data, Lan Xu

USF Tampa Graduate Theses and Dissertations

In epidemiologic and clinical studies, a relatively large number of biomarkers are repeatedly measured in patients over time, often associated with data on epidemiologic and clinical interest events. So, much attention is focused on developing the specific patterns of the longitudinal measurements, and the associations between those patterns and the time to a certain event, such as heart attack, diagnose of disease, time to transplantation, or death. In the last two decades, the research into joint modeling of longitudinal and time-to-event data has received a tremendous amount of attention.

Numerous researchers have proposed joint modeling approaches for a single longitudinal …

Data-Driven Analytical Modeling Of Multiple Myeloma Cancer, U.S. Crop Production And Monitoring Process, Lohuwa Mamudu

USF Tampa Graduate Theses and Dissertations

Globally, cancer disease is a major health issue causing a lot of deaths. The duration of time an individual diagnosed with a particular type of cancer survives has become a major area of research concern. The Kaplan Meier and Cox Proportional Hazard (Cox-PH) model have been a traditionally used method for survival analysis of cancer data. These techniques of cancer survival analysis are developed from nonparametric and semi-parametric approaches, respectively, which are not as robust as a parametric approach. In this dissertation, we proposed a new method of cancer survival analysis based on a parametric approach using multiple myeloma cancer …

Combination Of Time Series Analysis And Sentiment Analysis For Stock Market Forecasting, Hsiao-Chuan Chou

USF Tampa Graduate Theses and Dissertations

The goal of this research is to build a model to predict trend of financial asset price using sentiment from news headlines and financial indicators of the asset. Objective of the model is to conclude good results but also to minimize the difference between predicted values and actual values. Unlike previous approaches where the sentiments are usually calculated into score, we focus on combination of word embedding of news and financial indicators due to nonavailability of sentiment lexicon.

One idea is that the sentiment of news headline should have impact on financial asset val- ues. In other words, it would …

The General Psychopathology Factor (P) From Adolescence To Adulthood: Disentangling The Developmental Trajectories Of P Using A Multi-Method Approach, Alexandria M. Choate

USF Tampa Graduate Theses and Dissertations

Considerable attention is directed towards studying co-occurring psychopathology through the lens of a general factor (p-factor). However, the developmental trajectories and stability of the p-factor have yet to be fully understood. Study 1 first examined the explanatory power of dynamic mutualism theory — an alternative framework positing the p-factor to be a product of lower-level symptom interactions rather than the inherent cause of them. Predictions of dynamic mutualism were tested using three distinct statistical approaches including: longitudinal bifactor models, random-intercept cross-lagged panel models (RI-CLPMs), and network models. Next, given prior suggestions that borderline personality disorder (BPD) could be a marker …

Multimodal Data Fusion And Attack Detection In Recommender Systems, Mehmet Aktukmak

USF Tampa Graduate Theses and Dissertations

The commercial platforms that use recommender systems can collect relevant information to produce useful recommendations to the platform users. However, these sources usually contain missing values, imbalanced and heterogeneous data, and noisy observations. Such characteristics render the process of exploiting the information nontrivial, as one should carefully address them during the data fusion process. In addition to the degenerative characteristics, some entries can be fake, i.e., they can be the outcomes of malicious intents to manipulate the system. These entries should be eliminated before incorporation to any recommendation task. Detecting such malicious attacks quickly and accurately and then mitigating them …

Numerical Study Of Gap Distributions In Determinantal Point Process On Low Dimensional Spheres: L-Ensemble Of O(N) Model Type For N = 2 And N = 3, Xiankui Yang

USF Tampa Graduate Theses and Dissertations

Poisson point process is the most well-known point process with many applications. Unlike Poisson point process, which is the random set of non-intersecting points, determinantal point process refers to certain class of point processes where the points tend to interact with each other. The interaction often leads to more uniformly distributed points compared to those in Poisson point process.

In this article, we study the gap distribution of certain class of determinantal point process, L-ensemble of O(n) model type, and compare the distribution with the ones from the other known determinantal point process that appears in random matrices. Our numerical …

Bayesian Reliability Analysis For Optical Media Using Accelerated Degradation Test Data, Kun Bu

USF Tampa Graduate Theses and Dissertations

ISO (the International Organization for Standardization) 10995:2011 is the inter-national standard providing guidelines for assessing the reliability and service life of optical media, which is designed to be highly reliable and possesses a long lifetime. A well-known challenge of reliability analysis for highly reliable devices is that it is hard to obtain sufficient failure data under their normal use conditions. Accelerated degradation tests (ADTs) are commonly used to quickly obtain physical degradation data under elevated stress conditions, which are then extrapolated to predict reliability under the normal use condition. This standard achieves the estimation of the lifetime of recordable media, …

Identification Of Patterns And Disruptions In Ambient Sensor Data From Private Homes, Yan Wang

USF Tampa Graduate Theses and Dissertations

The world’s population is rapidly aging and the increasing demand for home and health care services from this aging population brings unprecedented challenges to the economy and society. Ambient-assisted smart homes, residences equipped with ambient sensors to monitor the resident’s daily activities in a continuous and unobtrusive way, present great potential to manage the growing care service needs of this older population segment, and enable them to age-in-place.

Despite growing research, using ambient sensor data from private homes to monitor daily activities, health and wellness still faces significant challenges. To study ambient sensor data from private homes where annotated data …

Predictive Validity Of Standards-Based And Curriculum-Embedded Assessments For Predicting Readiness At Kindergarten Entry, Elizabeth Ashton Decamilla

USF Tampa Graduate Theses and Dissertations

As with traditional K-12 educational settings, early childhood assessments have been a primary source of information determining whether early educational experiences have promoted children’s readiness to start school in kindergarten. The level of use of Kindergarten Entry Assessments (KEAs) has become more wide-spread to establish levels of school readiness at kindergarten entry.

This quantitative, correlational study of children in schools that have blended Head Start/Voluntary Prekindergarten funded programs examined the predictive relationships between the independent variables (i.e., VPK Assessments and Teaching Strategies GOLD) and the dependent variable of kindergarten readiness, as measured by the Work Sampling System™ (WSS). Additionally, the …

Exploration Of Factors Associated With Perceptions Of Community Safety Among Youth In Hillsborough County, Florida: A Convergent Parallel Mixed-Methods Approach, Yingwei Yang

USF Tampa Graduate Theses and Dissertations

Introduction: Youth perceived safety is not only linked to crime and violence in a neighborhood but is also associated with health risk behaviors and certain neighborhood characteristics. The purpose of this mixed-methods study was to measure the co-occurring effects of individual and community risk factors by conducting a secondary data analysis using structural equation modeling (SEM) and to explore reasons for youth feeling safe/unsafe in their community using photovoice methodology.

Methods: Syndemic theory/model served as the theoretical framework to guide this mixed-methods study with a convergent parallel design. The quantitative strand (first manuscript) utilized an existing dataset collected from middle …

Bayesian Reliability Analysis Of The Power Law Process And Statistical Modeling Of Computer And Network Vulnerabilities With Cybersecurity Application, Freeh N. Alenezi

USF Tampa Graduate Theses and Dissertations

As most of mankind now lives in an era of high dependence on multiple technologies and complex systems to store and manage sensitive information, researchers are constantly urged to obtain and improve measurements and methodologies that have the ability to evaluate systems reliability and security. The objectives of the present dissertation are to improve the Bayesian reliability estimation of a software package where the Power Law Process, also known as Non-Homogeneous Poisson Process, is the underlying failure model and to develop a set of statistical models evaluating computer operating systems vulnerabilities. Furthermore, we develop a reliability function of a computer …

Gradient Boosting For Survival Analysis With Applications In Oncology, Nam Phuong Nguyen

USF Tampa Graduate Theses and Dissertations

Cancer is one of the most deadly diseases that the world has been fighting against over decades. An enormous number of research has been conducted, via a wide scale of approaches, raging from genetic analysis to mathematical modeling. Survival analysis is a well-performed methodology frequently used to estimate the survival probability of a patient. Although there has been a large number of methods for survival analysis, efficient exploration of a high-dimensional feature space has been challenging due to its computational cost and complexity. This thesis adapts the component-wise gradient boosting algorithms for cancer survival analysis, and also proposes a new …

Fractional Random Weighted Bootstrapping For Classification On Imbalanced Data With Ensemble Decision Tree Methods, Sean Charles Carter

USF Tampa Graduate Theses and Dissertations

Ensemble methods are commonly used for building predictive models for classification. Models that are unstable to perturbations in the training set, such as the decision tree, often see considerable reductions in error when grouped, using bootstrapped resamples of the training data to train many models. The non-parametric bootstrap, however, has limited efficacy when used on severely imbalanced data, especially when the number of observations of one or more classes is exceptionally small. We explore the fractional random weighted bootstrap, which randomly assigns fractional weights to observations, as an alternative resampling pro cedure in training machine learning ensembles, particularly decision tree …

Probabilistic Modeling Of Democracy, Corruption, Hemophilia A And Prediabetes Data, A. K. M. Raquibul Bashar

USF Tampa Graduate Theses and Dissertations

Parametric analysis of any real-world data is the most powerful tool to characterize the probabilistic behavior in social, economic, medical, epidemiological, and other areas of study. In the present study, we identify the theoretical Probability Distribution Function(PDF) for Democracy Index Scores (DIS) from the Economist Intelligence Unit (EIU) database and estimate the maximum likelihood estimates of the theoretical PDFS. We also identify the individual PDFs for each of the clusters, Full Democracy, Flawed Democracy, Hybrid Regime, and Authoritarian Regime defined by the Economist Intelligence Unit (EIU).

A statistical model is a convenient instrument to predict the future value of any …

Statistical Learning Of Biomedical Non-Stationary Signals And Quality Of Life Modeling, Mahdi Goudarzi

USF Tampa Graduate Theses and Dissertations

Statistical learning is a set of tools for modeling and understanding complex datasets. It is a recently developed area in statistics and blends with parallel developments in computer science and, in particular, machine learning.

The classification of biomedical non-stationary signals such as Electroencephalogram (EEG) is always a challenging problem due to their complexity. The low spatial resolution on the scalp, curse of dimensionality, poor signal-to-noise ratio are disadvantages of working with biomedical signals. EEG signals are unstructured data which needs preprocessing steps to extract informative features which are measurable and predictive. In the first two chapters of this dissertation, EEG …