Digital Commons Network^™

Designing For Deployable, Secure, And Generic Machine Learning Systems, Li-Yun Wang

Dissertations and Theses

Machine learning systems have catalyzed numerous image-centric applications owing to the significant achievements of machine learning algorithms and models. While these systems have showcased the efficacy of machine learning models, certain challenges persist, such as machine learning system design and security vulnerabilities inherent in deep neural networks. Moreover, the deployment of deep neural network models remains a significant hurdle. This dissertation introduces a multimedia prototyping framework tailored for visual analytical applications, improving the reusability of video analysis software tools with minimal performance overhead. Furthermore, we present novel image-processing techniques designed to bolster the robustness of deep neural networks and propose …

Developing A Sql Injection Exploitation Tool With Natural Language Generation, Kate Isabelle Boekweg

Theses and Dissertations

Websites are a popular tool in our modern world, used daily by many companies and individuals. However, they are also rife with vulnerabilities, including SQL injection (SQLI) vulnerabilities. SQLI attacks can lead to significant damage to the data stored within web applications and their databases. Due to the dangers posed by these attacks, many countermeasures have been researched and implemented to protect websites against this threat. Various tools have been developed to enhance the process of detecting SQLI vulnerabilities and active SQLI attacks. Many of these tools have integrated machine learning technologies, aiming to improve their efficiency and effectiveness. Penetration …

Driver’S Accident Behavioral Analytics Using Ai, Mohamad Amin Obaid

Theses

This comprehensive dissertation constitutes a significant contribution to the ongoing global discourse on road safety. Through a judicious utilization of advanced data analysis techniques, with a particular emphasis on machine learning applications, this research endeavors to address and bridge crucial gaps in our comprehension of multifaceted aspects related to road safety. Specifically, the study aims to delve into the intricacies of accident severity factors, driver characteristics, vehicle attributes, and the complex dynamics of road conditions. By systematically exploring these dimensions, the research endeavors to unearth more nuanced and precise relationships that influence accident outcomes. Moreover, a particular focus is dedicated …

Development Of A Molten Salt Reactor Design And Design Framework For Overall Reactor Performance Prediction And Optimization, Andrew William Larsen

Theses and Dissertations

This dissertation introduces a novel approach to nuclear reactor design, emphasizing molten salt reactors (MSRs) to address the global need for safer, more efficient, and sustainable energy sources. It confronts the inherent challenges of nuclear fission reactors--proliferation risks, cost efficiency, waste management, and passive accident prevention--by developing an innovative MSR model and a machine learning (ML)-based optimization framework. The molten salt microreactor (MSMR), devised by researchers at Brigham Young University, utilizes thermally conductive plates for efficient heat dissipation during operational anomalies, ensuring safety within operational limits. This design was rigorously tested using STARCCM+ for fluid dynamics modeling and RELAP5-3D for …

Sign-A-Mander: A Mobile App That Enhances Asl Learning With Computer Vision, Sandrine Adap

Honors Theses

Several machine learning researchers have developed algorithms recognizing American Sign Language (ASL), but few have applied the algorithms to real-world situations, such as with portable ASL learning applications. This project develops a beta version of a mobile application designed to allow beginner ASL learners to practice basic ASL vocabulary and receive feedback about their signing accuracy. Building on Dongxu Li et al.’s I3D sign language recognition algorithm and 2000-word dataset, the app seeks to determine whether the I3D algorithm can sufficiently recognize a user’s motions when recorded from a mobile device and accurately classify whether or not the user signed …

Academic Detailing As A Health Information Technology Implementation Method: Supporting The Design And Implementation Of An Emergency Department-Based Clinical Decision Support Tool To Prevent Future Falls, Hanna J Barton, Apoorva Maru, Margaret A Leaf, Daniel J Hekman, Douglas A Wiegmann, Manish N Shah, Brian W Patterson

Faculty, Staff and Student Publications

BACKGROUND: Clinical decision support (CDS) tools that incorporate machine learning-derived content have the potential to transform clinical care by augmenting clinicians' expertise. To realize this potential, such tools must be designed to fit the dynamic work systems of the clinicians who use them. We propose the use of academic detailing-personal visits to clinicians by an expert in a specific health IT tool-as a method for both ensuring the correct understanding of that tool and its evidence base and identifying factors influencing the tool's implementation.

OBJECTIVE: This study aimed to assess academic detailing as a method for simultaneously ensuring the correct …

Integrated Organizational Machine Learning For Aviation Flight Data, Michael J. Pritchard, Austin T. Walden, Paul J. Thomas

Journal of Aviation/Aerospace Education & Research

Increased availability of data and computing power has allowed organizations to apply machine learning techniques to various fleet monitoring activities. Additionally, our ability to acquire aircraft data has increased due to the miniaturization of small form factor computing machines. Aircraft data collection processes contain many data features in the form of multivariate time series (continuous, discrete, categorical, etc.) which can be used to train machine learning models. Yet, three major challenges still face many flight organizations: 1) integration and automation of data collection frameworks, 2) data cleanup and preparation, and 3) developing an embedded machine learning framework. Data cleanup and …

Evaluating The Effectiveness Of Dormant Season Grazing To Reduce Cheatgrass And Its Effect On Biodiversity, Curtis L. Drake

Theses and Dissertations

Invasion of cheatgrass (Bromus tectorum L.) into sagebrush steppe communities threatens the survival of these at-risk ecosystems. One tool available to combat cheatgrass invasion is dormant season grazing. This study investigates the effectiveness of dormant season cattle grazing to reduce cheatgrass abundance in two high-elevation sagebrush steppe communities in southeastern Idaho, as well as the impact of dormant season grazing on plant species biodiversity. Our experiment utilized a randomized complete block design with four grazing treatments – ungrazed, fall-grazed, spring-grazed, and a combination of spring- and fall-grazed. We used an ANOVA standard least squares procedure to determine differences in biodiversity …

Application Of High-Deflection Strain Gauges To Characterize Spinal-Motion Phenotypes Among Patients With Clbp, Spencer Alan Baker

Theses and Dissertations

Chronic low back pain (CLBP) is a nonspecific and persistent ailment that entails many physiological, psychological, social, and economic consequences for individuals and societies. Although there is a plethora of treatments available to treat CLBP, each treatment has varying efficacy for different patients, and it is currently unknown how to best link patients to their ideal treatment. However, it is known that biopsychosocial influences associated with CLBP affect the way that we move. It has been hypothesized that identifying phenotypes of spinal motion could facilitate an objective and repeatable method of determining the optimal treatment for each patient. The objective …

Context-Aware Affective Behavior Modeling And Analytics, Md Taufeeq Uddin

USF Tampa Graduate Theses and Dissertations

Affective computing (AC) is a sub-domain of AI that has the potential to assist people by assessing mental states and making appropriate recommendations to patients, loved ones, caregivers, and domain experts. Humans usually produce an enormous amount of data (such as face videos) every day. One of the major challenges for affective computer vision is to efficiently deal with high volumes of data to facilitate automated model development. To cope with this challenge, we developed computer vision algorithms that measure the expressivity of the human face from video data. More precisely, the developed algorithms can map complex affect information from …

Preparing And Teaching Data Science Courses, Dale E. Parson

Computer Science and Information Technology Faculty

Preparing and teaching realistic data science courses requires labor-intensive preparation and course delivery. It is not enough to download data and push buttons on machine learning tools. First, there must be a human expert available in the problem domain to supply data and evaluate work. Without a human expert to provide information that is missing or incorrect in archived data, the tendency is to take the output of machine learning algorithms using potentially faulty data on faith. Second, any real-world data requires custom scripts for correcting invalid values, creating derived attributes, and formatting data for analysis. Then comes the analysis, …

Forecasting Future Research Trends In The Construction Engineering And Management Domain Using Machine Learning And Social Network Analysis, Gasser G. Ali, Islam H. El-Adaway, Muaz O. Ahmed, Radwa Eissa, Mohamad Abdul Nabi, Tamima Elbashbishy, Ramy Khalef

Civil Engineering Faculty Publications

Construction Engineering and Management (CEM) is a broad domain with publications covering interrelated subdisciplines and considered a key source of knowledge sharing. Previous studies used scientometric methods to assess the current impact of CEM publications; however, there is a need to predict future citations of CEM publications to identify the expected high-impact trends in the future and guide new research efforts. To tackle this gap in the literature, the authors conducted a study using Machine Learning (ML) algorithms and Social Network Analysis (SNA) to predict CEM-related citation metrics. Using a dataset of 93,868 publications, the authors trained and tested two …

Using Data Analytics To Predict Government Employee Turnover, Aisha Alshamsi

Theses

Employee turnover is a significant problem in many organizations, resulting in increased costs, decreased productivity, and decreased morale. The aim of this research is to predict employee turnover using data analysis algorithms and develop a predictive model that can be used to identify employees who are at risk of leaving the organization. This study preprocesses data on employee turnover rates, employee demographics, employee satisfaction, and other relevant factors, and uses data analysis machine learning algorithms such as Logistic regression, Decision tree, and XGBoost to analyze the data and develop a predictive model. The result of the study helps impact on …

Classification Of Lakebed Geologic Substrate In Autonomously Collected Benthic Imagery Using Machine Learning, Joseph K. Geisz, Phillipe A. Wernette, Peter C. Esselman

Michigan Tech Publications

Mapping benthic habitats with bathymetric, acoustic, and spectral data requires georeferenced ground-truth information about habitat types and characteristics. New technologies like autonomous underwater vehicles (AUVs) collect tens of thousands of images per mission making image-based ground truthing particularly attractive. Two types of machine learning (ML) models, random forest (RF) and deep neural network (DNN), were tested to determine whether ML models could serve as an accurate substitute for manual classification of AUV images for substrate type interpretation. RF models were trained to predict substrate class as a function of texture, edge, and intensity metrics (i.e., features) calculated for each image. …

Ai Is A Viable Alternative To High Throughput Screening: A 318-Target Study, The Atomwise Aims Program

Faculty, Staff and Student Publications

High throughput screening (HTS) is routinely used to identify bioactive small molecules. This requires physical compounds, which limits coverage of accessible chemical space. Computational approaches combined with vast on-demand chemical libraries can access far greater chemical space, provided that the predictive accuracy is sufficient to identify useful molecules. Through the largest and most diverse virtual HTS campaign reported to date, comprising 318 individual projects, we demonstrate that our AtomNet® convolutional neural network successfully finds novel hits across every major therapeutic area and protein class. We address historical limitations of computational screening by demonstrating success for target proteins without known binders, …

Predictive Policing, Emily Musgrove

Richard T. Schellhase Essay Prize in Ethics

Currently, machine learning and artificial intelligence are still somewhat new to the world of law enforcement, however, they appear to be rapidly gaining traction for their usefulness in a variety of manners. In this paper, I will examine how technically useful machine learning-based predictive policing is, the ethical conundrums that come with its practical uses, and how to mitigate the risks of its use.

Enhancing Rotating Machinery Fault Diagnosis: A Dual-Head Attention Mechanism In Deep Learning Neural Networks, Qing Snyder

Dissertations

Rotating machinery is crucial to production efficiency and safety in manufacturing industries for an extended time. Ensuring machinery reliability necessitates effective diagnostic systems, particularly for rotating bearings, the key components of such equipment. Fault diagnosis in rotating machinery is essential to prevent failures and minimize downtime, thereby playing an important role in industrial operations. The application of advanced neural network techniques in industry has risen recently. Among these, attention-based neural networks, especially the Transformer models, are originally noteworthy for their sequential data handling capability. This research delves into attention-based algorithms for rotating machinery fault diagnosis, signifying a substantial advancement in …

Predictive Analytics Of Road Traffic Incidents, A Machine Learning Approach, Maryam Essa Jaji

Theses

Traffic accidents rank among the world's most serious concerns due to the high number of deaths, injuries, and fatalities as well as the enormous financial losses they cause every year. Road travel is a necessary component of modern civilization, but because of the rise in traffic accidents, it costs the world economy billions of dollars and over a million deaths annually. Road accidents can be caused by a variety of elements. It can be possible to take action to lessen the severity and extent of the effects if these elements are better recognized and predicted. The goal of this project …

Advanced Machine Learning Approaches For Predicting Mental Health Disorders Following Long Covid Diagnosis, Manoj Purohit

Master's Theses (2009 -)

After the global spread of COVID-19, the enduring effects of Long COVID and its health implications have emerged as a significant global issue, affecting people worldwide. The lingering symptoms post a COVID-19 infection can significantly affect individuals who had previously contracted the virus, exerting considerable influence over their mental well-being. Prolonged recuperation associated with Long COVID has been connected with the emergence of symptoms such as depression and anxiety, all of which can have adverse effects on emotional health. This project delves into an in-depth analysis of healthcare data pertaining to Long COVID from the Froedtert Health (FH) Medical System …

Leveraging Ai And Machine Learning To Develop And Evaluate A Contextualized User-Friendly Cough Audio Classifier For Detecting Respiratory Diseases: Protocol For A Diagnostic Study In Rural Tanzania, Kahabi Isangula, Rogers John Haule

School of Nursing & Midwifery, East Africa

Background:

Respiratory diseases, including active tuberculosis (TB), asthma, and chronic obstructive pulmonary disease (COPD), constitute substantial global health challenges, necessitating timely and accurate diagnosis for effective treatment and management.

Objective:

This research seeks to develop and evaluate a noninvasive user-friendly artificial intelligence (AI)–powered cough audio classifier for detecting these respiratory conditions in rural Tanzania.

Methods:

This is a nonexperimental cross-sectional research with the primary objective of collection and analysis of cough sounds from patients with active TB, asthma, and COPD in outpatient clinics to generate and evaluate a noninvasive cough audio classifier. Specialized cough sound recording devices, designed to be …

Robust Machine Learning Under Vulnerable Cyberinfrastructure And Varying Data Quality, Sergei Chuprov

Theses

In our study, we investigate Machine Learning (ML) application robustness in ML Integrated with Network (MLIN) systems. We consider MLIN as an integration of three major components and interrelations between them: data sources, network facilities, and ML application. In contrast to the conventional approaches that focus on each separate component, we concentrate on their interrelationships, and consider them from the system integration perspective. We formulate our primary goal as to develop methods and tools aimed at assuring ML application performance towards Data Quality (DQ) variation in MLIN through improving ML application robustness. We examine the prior work on ML robustness …

Learning To Outsmart A Crisis: A Strategic Management View On Managing Aviation Crises With Machine Learning, Kok Yew Tan

Dissertations and Theses Collection (Open Access)

In strategic management literature, a crisis is public, unexpected, interferes with the norm, has negative impact on organizations, and generates widespread negative perceptions among stakeholders. Among crises, aviation crises are complex and present multiple vulnerabilities due to their scale, and the involvement of numerous stakeholders. With the high visibility of aviation crises due to the potential for mass injuries, deaths, and collateral damage, it could be seen as a full spectrum crisis – one of the worst among crises. The magnitude of each aviation crisis decides and allows the world to assess the efficacy of an airlines or aviation firm’s …

Ai-Assisted Reconfiguration Of Battery Packs For Cell Balancing To Extend Driving Runtime, Yuqin Weng, Cristinel Ababei

Electrical and Computer Engineering Faculty Research and Publications

State of charge (SoC) cell balancing is one of the most important roles of battery management systems (BMS). The performance and lifespan of a battery pack can be significantly degraded and reduced by the presence of imbalance in cells SoC. Recently, we have shown that using a machine learning driven battery pack reconfiguration technique based on a network of controllable switches, one can periodically change the battery pack topology to effectively achieve better cell SoC equalization. As a result, the driving runtime achieved with a better balanced battery pack is increased. In this paper, we build on these promising results …

Automated Glacier Classification In High Mountain Asia Using Machine Learning And A Random Forest Classifier, Victoria Elizabeth Halvorson

Dartmouth College Master’s Theses

High Mountain Asia (HMA) is home to the largest mass of glaciers and ice outside the north and south polar regions. HMA glaciers are projected to experience accelerated mass loss from higher greenhouse gas emissions through the end of the century. Many studies of glacier mass balance and mass loss in HMA obtain glacier area from the Randolph Glacier Inventory (RGI). However, the RGI is designed to show glacier area across the world that is accurate to the year 2000 and, as a result, is not an accurate representation of the current state of glacier area in HMA. Additionally, glacier …

Distinguishing Leukemic Cells Using Fractal Chromatin Patterns And Machine Learning, Abigail Gordhamer, Ryan Cordner, Paul Young

Library/Life Sciences Undergraduate Poster Competition 2024

One of the most important tests in the clinical laboratory is the Complete Blood Count, which involves identifying the white blood cells in a patient’s blood. The respective counts of the different white blood cell types correlate with various states of health and disease and are critical to diagnosing diseases such as leukemia. Leukemic cells (blasts) are considered especially difficult to distinguish, and it is of the upmost importance that these cells are identified correctly. To aid in the process of leukemic cell identification, we quantified fractal patterns in the chromatin of white blood cells and used the data to …

Using Chatgpt To Generate Gendered Language, Shweta Soundararajan, Manuela Nayantara Jeyaraj, Sarah Jane Delany

Conference papers

Gendered language is the use of words that denote an individual's gender. This can be explicit where the gender is evident in the actual word used, e.g. mother, she, man, but it can also be implicit where social roles or behaviours can signal an individual's gender - for example, expectations that women display communal traits (e.g., affectionate, caring, gentle) and men display agentic traits (e.g., assertive, competitive, decisive). The use of gendered language in NLP systems can perpetuate gender stereotypes and bias. This paper proposes an approach to generating gendered language datasets using ChatGPT which will provide data for data-driven …

Automated Seizure Detection Based On State-Space Model Identification, Zhuo Wang, Michael Sperling, Dale Wyeth, Allon Guez

Department of Neuroscience Faculty Papers

In this study, we developed a machine learning model for automated seizure detection using system identification techniques on EEG recordings. System identification builds mathematical models from a time series signal and uses a small number of parameters to represent the entirety of time domain signal epochs. Such parameters were used as features for the classifiers in our study. We analyzed 69 seizure and 55 non-seizure recordings and an additional 10 continuous recordings from Thomas Jefferson University Hospital, alongside a larger dataset from the CHB-MIT database. By dividing EEGs into epochs (1 s, 2 s, 5 s, and 10 s) and …

Utilizing Machine Learning Techniques For Accurate Diagnosis Of Breast Cancer And Comprehensive Statistical Analysis Of Clinical Data, Myat Ei Ei Phyo

USF Tampa Graduate Theses and Dissertations

Breast cancer represents a formidable malignancy, presenting a substantial threat to global health and individual well-being. Conventionally, it is widely held that the prognosis for breast cancer patients hinges predominantly upon the timing of diagnosis and the extent of cancer progression, typically delineated by its stage. However, emerging evidence from robust regression and machine learning analyses challenges this prevailing notion. The results indicate that survival months cannot be solely attributed to diagnosis and socio-economic factors. Instead, additional variables such as existing diseases and treatment complexities may contribute to the intricate landscape of breast cancer outcomes.

This research aims to delve …

Anti-Phishing Approach For Iot System In Fog Networks Based On Machine Learning Algorithms, Mahmoud Gad Awwad, Mohamed M. Ashour, El Said A. Marzouk, Eman Abdelhalim

Mansoura Engineering Journal

As the Internet of Things (IoT) continues to expand, ensuring the security and privacyِ of IoT systems becomes increasingly critical. Phishing attacks pose a significant threat to IoT devices and can lead to unauthorized access, data breaches, and compromised functionality. In this paper, we propose an anti-phishing approach for IoT systems in fog networks that leverages machine learning algorithms, including a .fusion with deep learning techniques We explore the effectiveness of eleven traditional machine learning algorithms combined with deep learning in detecting and preventing phishing attacks in IoT systems. By utilizing a diverse range of algorithms, we aim to enhance …

Gt-Ches And Dycon: Improved Classification For Human Evolutionary Systems, Joseph S. Johnson

Theses and Dissertations

The purpose of this work is to rethink the process of learning in human evolutionary systems. We take a sober look at how game theory, network theory, and chaos theory pertain specifically to the modeling, data, and training components of generalization in human systems. The value of our research is three-fold. First, our work is a direct approach to align machine learning generalization with core behavioral theories. We made our best effort to directly reconcile the axioms of these heretofore incompatible disciplines -- rather than moving from AI/ML towards the behavioral theories while building exclusively on AI/ML intuition. Second, this …