Data Science Commons^™

Applying Software Engineering Black-Box Methods For Testing Machine Learning Models, Timothy Elvira

Doctoral Dissertations and Master's Theses

This dissertation proposes researching an approach to incorporate and align Software black-box testing methods into Machine Learning (ML) applications, specifically in the context of computer vision models. Typically, testing methods within Software Engineering (SE) encompass a range of test types that assess levels of a software system, such as Unit, Integration, Functional, and System testing [1]. The testing spectrum offers two perspectives on the system: black-box, where the system’s code is hidden, and white-box, where the system's code is exposed for testing. Software Quality pairs testing with requirements, in a many-to-one relationship, to ensure proper validation of the software system. …

Surgical Versus Nonsurgical Management Of Civilian Craniocerebral Gunshot Injuries, Wesley Shoap, George Austin Crabill, Roboan Guillen, Kaleb Derouen, Jack Leoni, Zhide Fang, Berje Shammassian

School of Medicine Faculty Publications

Introduction: Craniocerebral gunshot wounds in the civilian population constitute a devastating subset of traumatic brain injuries (TBI). The aim of this study was to determine the association of mortality, intensive care unit length of stay (ICU LOS), and the Glasgow Outcome Scale Extended (GOS-E) among craniocerebral gunshot patients based on timing and type of intervention. Methods: The trauma database was queried for GSWH patients ages 15 and older who received neurosurgical intervention from January 1st 2016 to June 1st 2023. Operative notes were reviewed and patients were then divided into three groups; intracranial pressure monitor only with medical treatment (ICP), …

Spotify Recommender Using Content Filtering, Colin M. Anderson

SPARK Symposium Presentations

This project is a music recommender system that analyzes Spotify data to establish relationships between songs by analyzing their musical components. A user can receive recommendations through two methods. Firstly, the user can select a song from the database that they enjoy, and the system will provide them with a list of recommendations, as well as predict the genre of the song that they have entered. Secondly, the user can set personalized values for various musical features (i.e. energy, danceability). In either case, by selecting a number 1 to 5 recommendations that the user would like, they will get that …

Liver Tet1 Promotes Metabolic Dysfunction-Associated Steatotic Liver Disease, Hongze Chen, Muhammad Azhar Nisar, Joud Mulla, Xinjian Li, Kevin Cao, Shaolei Lu, Katsuya Nagaoka, Shang Wu, Peng Sheng Ting, Tung Sung Tseng, Hui Yi Lin, Xiao Ming Yin, Wenke Feng, Zhijin Wu, Zhixiang Cheng, William Mueller, Amalia Bay, Layla Schechner, Xuewei Bai, Chiung Kuei Huang

School of Public Health Faculty Publications

Global hepatic DNA methylation change has been linked to human patients with metabolic dysfunction-associated steatotic liver disease (MASLD). DNA demethylation is regulated by the TET family proteins, whose enzymatic activities require 2-oxoglutarate (2-OG) and iron that both are elevated in human MASLD patients. We aimed to investigate liver TET1 in MASLD progression. Depleting TET1 using two different strategies substantially alleviated MASLD progression. Knockout (KO) of TET1 slightly improved diet induced obesity and glucose homeostasis. Intriguingly, hepatic cholesterols, triglycerides, and CD36 were significantly decreased upon TET1 depletion. Consistently, liver specific TET1 KO led to improvement of MASLD progression. Mechanistically, TET1 promoted …

Immersive Mixed-Reality Anatomy Lab, Cassidy Norkett

Honors Theses

The Mixed Reality Heart Anatomy Lesson was created to address the WMU Homer Stryker M.D. School of Medicine’s need for an interactive and dynamic laboratory lesson. The application allows for a more efficient way for medical students to learn and comprehend the material. The design and development of a mixed reality application running on smart glasses provides students with a situation aware, fully engaged, and immersive learning experience. The Apple Vision Pro device allows medical students to complete their labs by merging physiology and anatomy concepts through 3D models, object detection, self-assessments, and an artificial intelligence chatbot. There can be …

Cost Modeling For State Longitudinal Data Systems, Benjamin Boer

Practitioner Toolkits and Resources

A tool to support the development and scaling of longitudinal data supports in state policymaking.

Differences In Covid-19 Deaths Amongst Cancer Patients And Possible Mediators For This Relationship, Leah Vaidya, Nubaira Rizvi, Xiao Cheng Wu, Lauren S. Maniscalco, Yong Yi, Augusto Ochoa, Qingzhao Yu

School of Public Health Faculty Publications

Previous research demonstrated Non-Hispanic Black populations experience higher COVID-19 mortality rates than Non-Hispanic White individuals. Additionally, cancer status is a known risk factor for COVID-19 death. While prior studies investigated comorbidities as exploratory variables in differences in COVID-19 hospitalization, none have explored their role in COVID-19-related deaths. This study aimed to evaluate whether Charlson Comorbidity Index (CCI) and subsequently, individual diseases are potential explanatory variables for this relationship. The analysis focused on Non-Hispanic Black and Non-Hispanic White cancer patients aged 20 or older, diagnosed between 2011 and 2019, who tested positive for COVID-19 from the start of pandemic through June …

Unlocking Precision Using K-Means++- Improved Genetic Algorithm-Radial Basis Function Neural Network: Data-Driven Evolution Of Smart Gloves For Gesture Recognition, Liang Xiao Ding, Kuan Way Chee, Hong Lü, Anand Paul, Jeonghong Kim, Jang Myung Lee

School of Public Health Faculty Publications

Human-computer interaction technologies have been used since the 1970s but have only gained growing popularity in recent years with new design paradigms. Ongoing research and development in gesture recognition systems with broad application prospects have focused on improving accuracy and real-time performance as well as the robustness of specific machine learning algorithms against environmental conditions. This paper addresses the accuracy enhancement of a novel Fifth Dimension Technologies data-glove-based gesture recognition system using a genetic-algorithm (GA)-trained k-means++-improved radial basis function (RBF) or GK-RBF neural network. First, we analyzed and modeled the sensor distribution in the data glove and proposed joint constraints …

Student Expectations And Outcomes In Virtual Vs In-Person Interprofessional Simulations: A Qualitative Analysis, Padmavathy Ramaswamy, Abbey M Bachmann, Tiffany Champagne-Langabeer, Chasisty L Gilder, Samuel E Neher, Jennifer L Swails

Faculty, Staff and Student Publications

Background: Health-related programs frequently integrate interprofessional education (IPE) into their training. The COVID-19 pandemic transitioned many IPE programs online, making it essential to assess student expectations and perceived learning outcomes across virtual simulations and in-person settings.

Methods: This qualitative study compared student expectations and self-reported outcomes across in-person and virtual case scenarios at a Texas health science center. Responses to open-ended questions from two data collection periods were analyzed using inductive coding and thematic analysis.

Results: Students from nursing, medicine, dentistry, public health, and informatics participated in each group. Three major themes emerged from this study: communication, teamwork, and …

Linking Water Quality And Climate Change To Long-Term Trends In Species Abundance In Norwalk Harbor, Viktoria Savatorova, Aidan Kieft, Nicole C. Spiller, Kasey Burns

Spora: A Journal of Biomathematics

This study examines the effects of environmental changes on fish populations in Norwalk Harbor, focusing on winter flounder (Pseudopleuronectes americanus), cunner (Tautogolabrus adspersus), northern pipefish (Syngnathus fuscus), and naked goby (Gobiosoma bosci) as examples of species responding to climate-related shifts. We analyze how water temperature, salinity, and dissolved oxygen correlate with fish abundance. To assess statistically significant differences in catch per unit effort (CPUE) across harbor regions, we applied the Kruskal-Wallis test followed by Dunn's post-hoc test. Seasonal variations in CPUE were examined by comparing monthly catch data for each species. K-means …

A Deep Sparse Capsule Network For Non-Invasive Blood Glucose Level Estimation Using A Ppg Sensor, Narmatha Chellamani, Saleh Ali Albelwi, Manimurugan Shanmuganathan, Palanisamy Amirthalingam, Emad Muteb Alharbi, Hibah Qasem Salman Alatawi, Kousalya Prabahar, Jawhara Bader Aljabri, Anand Paul

School of Public Health Faculty Publications

Diabetes, a chronic medical condition, affects millions of people worldwide and requires consistent monitoring of blood glucose levels (BGLs). Traditional invasive methods for BGL monitoring can be challenging and painful for patients. This study introduces a non-invasive, deep learning (DL)-based approach to estimate BGL using photoplethysmography (PPG) signals. Specifically, a Deep Sparse Capsule Network (DSCNet) model is proposed to provide accurate and robust BGL monitoring. The proposed model’s workflow includes data collection, preprocessing, feature extraction, and predictions. A hardware module was designed using a PPG sensor and Raspberry Pi to collect patient data. In preprocessing, a Savitzky–Golay filter and moving …

Precision Phenotyping For Curating Research Cohorts Of Patients With Unexplained Post-Acute Sequelae Of Covid-19, Alaleh Azhir, Jonas Hügel, Jiazi Tian, Jingya Cheng, Ingrid V Bassett, Douglas S Bell, Elmer V Bernstam, Maha R Farhat, Darren W Henderson, Emily S Lau, Michele Morris, Yevgeniy R Semenov, Virginia A Triant, Shyam Visweswaran, Zachary H Strasser, Jeffrey G Klann, Shawn N Murphy, Hossein Estiri

Faculty, Staff and Student Publications

BACKGROUND: Scalable identification of patients with post-acute sequelae of COVID-19 (PASC) is challenging due to a lack of reproducible precision phenotyping algorithms, which has led to suboptimal accuracy, demographic biases, and underestimation of the PASC.

METHODS: In a retrospective case-control study, we developed a precision phenotyping algorithm for identifying cohorts of patients with PASC. We used longitudinal electronic health records data from over 295,000 patients from 14 hospitals and 20 community health centers in Massachusetts. The algorithm employs an attention mechanism to simultaneously exclude sequelae that prior conditions can explain and include infection-associated chronic conditions. We performed independent chart reviews …

Evaluating The Meditation Practices And Barriers To Adopting Mindful Medicine Among Physicians, Tiffany Champagne-Langabeer, Chelsea G Ratcliff, Christine Bakos-Block, Francine Vega, Marylou Cardenas-Turanzas, Aila Malik, Radha Korupolu

Faculty, Staff and Student Publications

Background: Chronic pain affects over 25% of U.S. adults and is a leading cause of disability. Mindfulness meditation (MM) is a nonpharmacologic approach to manage pain and improve well-being. Despite mounting evidence supporting its efficacy, MM remains underutilized in medical practice. Understanding physicians' engagement with MM and the barriers they face can inform strategies for integration into clinical care. This study assessed physicians' attitudes toward MM, including barriers to practice and their likelihood of recommending it to patients.

Methods: A cross-sectional survey of U.S. physicians was conducted from April to July 2024. Participants provided information on demographics, health struggles, and …

Logiclm: Robust Application Of Large Language Models With Logic Programming For Data Analytics, Evgeny Skvortsov, Shayan Mirjafari, Ojaswa Garg, Yilin Xia, Shaun Bowers, Bertram Ludäscher

Computer Science Faculty Scholarship

We present LogicLM, an OLAP-style interactive data analysis system that leverages large language models (LLMs) and is configured using Logica, an enhanced logic programming language with aggregation support that compiles to SQL. LogicLM uses an LLM to translate natural language queries by end users into executable code for automatically generating data visualizations. For each natural-language query, LogicLM provides a verifiable OLAP-based configuration that users can view and modify to help ensure results are reliable and accurate. This configuration, with measures, dimensions, and filters defined as logical predicates, offers a unified and user-friendly approach to naturallanguage data exploration, while keeping end …

Leveraging Large Language Models For Knowledge-Free Weak Supervision In Clinical Natural Language Processing, Enshuo Hsu, Kirk Roberts

Faculty, Staff and Student Publications

The performance of deep learning-based natural language processing systems is based on large amounts of labeled training data which, in the clinical domain, are not easily available or affordable. Weak supervision and in-context learning offer partial solutions to this issue, particularly using large language models (LLMs), but their performance still trails traditional supervised methods with moderate amounts of gold-standard data. In particular, inferencing with LLMs is computationally heavy. We propose an approach leveraging fine-tuning LLMs and weak supervision with virtually no domain knowledge that still achieves consistently dominant performance. Using a prompt-based approach, the LLM is used to generate weakly-labeled …

Unraveling Genetic Links Between Diabetes And Heart Failure-A Machine Learning Approach, Sunakhi Sahoo, Marzieh Ayati

Research Symposium

Background: Diabetic heart failure (DHF) is defined as a chronic and progressive disease which is associated with both diabetes and heart failure (HF). Even though there have been many developments in the knowledge of these diseases, there is still much to learn about the genetic crossovers between the two. In this study, we identified genes that are associated with diabetic heart failure and heart failure by using gene expression data from patients with DHF, HF, and a control group of patients who died of natural causes. We sought to identify genes that had altered expression levels which could possibly play …

Urban Heat Dynamics In Pune: The Influence Of Land Cover And Local Climate, Arpit Tiwari, Preethi Nanjundan, Ravi Ranjan Kumar, Ananya Karmakar, Satyaban Bishoyi Ratna

Northeast Journal of Complex Systems (NEJCS)

Urban areas with high population density and extensive infrastructure development have been experiencing an increasing strain on the local heat budget, leading to a surge in heat-related illnesses and discomfort. This study examined the impact of climate and land use as heat islands in Pune, India, from 2012 to 2023 at six different locations representing varying degree of urbanization. Satellite land cover observations revealed that 55.17% of the total area was urbanized in the city itself, which was limited to 44.8% in 2012. This urbanization has significantly impacted the increasing tendency of maximum temperature (T_max; 0.13℃ to 1.63℃ …

Analysis Of Systematic Trade-Offs Between Military And Healthcare Expenditure Alongside Gdp Growth Of Select Asian And Western Exporting Economies In The 21st Century, Rahul Balamurugan, Carlos Gershenson, Preethi Nanjundan, Hiroki Sayama

Northeast Journal of Complex Systems (NEJCS)

This study explores the complexity in the trade-offs between military expenditure, healthcare expenditure, and GDP growth across select Asian nations and major weapon-exporting countries, examining how nations allocate finite resources between national security and human well-being over the past two decades. Using a systems science approach, the research integrates Granger causality testing to analyze temporal and directional relationships among GDP growth, military expenditure, and healthcare expenditure, uncovering their dynamic interdependencies. The methodology includes trend and slope analysis, Granger causality testing, outlier detection, and clustering to identify heterogeneity in resource allocation strategies. Developed, weapon-exporting nations exhibit complementary trends, with strong causality …

Forecasting Army Recruiting Mission Distribution, Maxwell A. Deihl

Theses and Dissertations

The Army’s recruiting landscape has changed markedly in recent years, raising questions about whether forecasting methods of Army contracts remain robust. This thesis recreates the presented models in Joshua McDonald’s 2015 thesis. It replicates and evaluates the models with updated data (2018–2023) to assess their current validity and compare them to novel alternative approaches, such as simpler regression models or neural networks. While the 2015 model remains a valuable baseline, results suggest that either refining its variables or adopting alternative methods can improve predictive accuracy and interpretability. Ultimately, the United States Army Recruiting Command has many options regarding how it …

Collaborative Ai: Oer Materials For Exploring Ai As A Partner Rather Than A Tool, David Smith

Open Educational Resources

The Collaborative AI Open Educational Resource (OER) explores how artificial intelligence can act as a creative and analytical collaborator rather than a tool. Centered on the Balanced Blended Space (BBS) framework and the philosophy of the Center for Holistic Integration (CHI), the OER includes curriculum materials, theoretical models, and live research environments. It offers an interesting approach to blending physical, virtual, and conceptual spaces through shared human–AI agency and invites ongoing participation in interdisciplinary meta-projects.

Analyzing U.S. Army Recruiter Productivity Through Cohort Differentiation And Behavioral Tendency Composition, Mary M. Bates

Theses and Dissertations

This research analyzes differences among aggregate achievements of U.S. Army recruiting cohorts, determines which behavioral tendencies are indicative of performance level, and investigates aggregate behavioral composition with cohort achievement. Analyses require implementation of OLS regression, ANOVA, Tukey’s Test, Mann-Whitney U test, Holm-Bonferroni adjustment, XGBoost decision tree, logistic regression, and the Kolmogorov-Smirnov test. The results show insignificant achievement differences among cohorts and weak yet prevalent abilities of select measures of behavioral tendencies to indicate recruiter performance.

Class Imbalance: A Landscape Of Classification Models, Joshua L. Edmonds

Theses and Dissertations

Class imbalance poses significant challenges in machine learning classification. This study evaluates the performance of seven models (ANN, k-Means, kNN, LDA, LR, SVM, XGBoost) across multiple imbalance levels (10\%, 5\%, 1 \%, 0.5\%) and investigates the effectiveness of sampling techniques (Undersampling, SMOTE, SMOTE-ENN). ANOVA results confirm that model choice is the most critical factor, with XGBoost and SVM demonstrating superior robustness. SMOTE improves recall but reduces precision, while undersampling generally degrades overall performance. While significant, imbalance levels do not play a critical role in model effectiveness.

Tracking News Narratives: Topic Modeling, Sentiment, And Media Coverage Patterns, Alexandria G. Lai

Theses and Dissertations

This study introduces a novel content-driven influence measurement framework, built around a custom influence formula that integrates Non-negative Matrix Factorization (NMF) topic modeling, sentiment analysis, and influence metrics to analyze media narratives over time. Applied to news coverage of the 2020 U.S. presidential election and the COVID-19 pandemic, the framework identifies key topics, sentiment patterns, and influential sources. Results demonstrate its ability to distinguish between transient political controversies and sustained public health discourse while capturing shifts in media influence. While effective, refinements in topic separation, sentiment analysis, and temporal weighting could enhance adaptability. This study highlights the novel influence formula …

A Reinforcement Learning Approach For Maneuvering And Firing Decisions In Sead Operations, Nathaniel Garcia

Theses and Dissertations

The integration of automated processes in defense continues to expand, enhancing the lethality of military forces. Artificial intelligence accelerates decision-making cycles, removes the constraints of human-operated hardware, and improves coordination by enabling seamless integration across multiple systems. Suppression of Enemy Air Defenses (SEAD) missions are critical to the United States (U.S.) military, as they neutralize hostile air defense systems, ensuring air superiority and enabling safe and effective operations for aircraft in contested environments. Therefore, it is necessary to pair emerging autonomous capabilities with an important mission set in defense. This research investigates the Autonomous Unmanned Air-to-Ground Strike (AUAGS) problem, modeling …

Symbology Detection And Numerical Recognition For T-38 Heads-Up Display Recordings, Ben T. Hepner

Theses and Dissertations

The extraction of symbology and numerical data from the T-38 Heads-Up Display (HUD) enhances post-flight analysis and supports real-time decision-making. This research develops a deep learning pipeline using YOLO-based object detection and Optical Character Recognition (OCR) to analyze HUD video data. Model evaluations showed mAP0.5:0.95 ranging from 0.422 (YOLOv11m, hard test set) to 0.696 (YOLOv8m, medium test set), demonstrating robust symbology detection. Numeric detection performed well (mAP_0.5:0.95 = 0.764), but OCR struggled with glare and resolution limitations, achieving a recognition accuracy of 17.35%. These results validate deep learning for HUD data extraction but highlight the need for improved robustness …

Practical Estimation Of Action-Generation Mechanisms In Repeated Games, Vladimer Kellachow Iii

Theses and Dissertations

The goal of this research is to gain insight into how players of a game learn their strategy during the course of repeated play. The study employs the Experience Weighted Attraction (EWA) model, developed by Dr. Colin F. Camerer and Dr. Teck-Hua Ho, as the foundational behavioral framework. Using historic observed strategy decisions, the parameter values that define an opponent’s learning process are updated using various inference methods.

Machine Learning With Flight Data Recorder Data For Flight Fuel Consumption Predictions, Adam C. Levandowski

Theses and Dissertations

This study applies advanced Machine Learning (ML) to Flight Data Recorder (FDR) data for fuel consumption predictions. It explores feature engineering, model selection, and Hyper-Parameter Optimization (HPO) across all flight phases. Baseline models like Ordinary Least Squares (OLS) regression, Multi- Layer Perceptrons (MLPs), and decision trees are compared to Deep Neural Networks (DNNs), Recurrent Neural Networks (RNNs) with Gated Recurrent Unit (GRU) layers, and XGBoost. Results analyze segmentation strategies, tailored features, and model performance. A counterfactual analysis compares ML models to operational fuel predictions, demonstrating their deployment potential. Findings establish a foundation for future ML-driven advancements in aviation fuel optimization.

Data Lakehouse And Machine Learning Pipeline For Aircraft Fuel Efficiency Experimentation, Skyler G. Kepley

Theses and Dissertations

Fuel efficiency is crucial for the U.S. Air Force, impacting mission success, aircraft performance, and cost savings. This study presents an information system that integrates flight and maintenance data using a data lakehouse. It automates ingestion, enrichment, and predictive modeling, leveraging AutoML for optimization and SHAP for transparency. A case study on C-130J aircraft shows that optimizing D Check cycles can save 11.52 pounds of fuel per flight hour. These findings highlight the effectiveness of data-driven decision-making in aviation, offering a scalable, automated solution for improving fuel efficiency and reducing costs.

Topical Text Segmentation For Stream Of Consciousness Writing, Yuwei Lu, Ryan Benton

Shelby Hall Graduate Research Forum Posters

Stream of consciousness writing has a long history, including novelists James Joyce and Virginia Woolf. However, there has been little work done in automated and semi-automated analysis of such writing, which is the focus of this work. We plan to divide real streams of consciousness writing into distinct topical units and then capture different momentary meaningful topics from these units. By doing this, researchers and readers could gain a more nuanced understanding of the narrative structure and thematic elements. In addition, it would also support applications in fields like psychology and linguistics, where understanding thought processes and narrative structures is …

Proximal Policy Optimization Applied To The Beyond Visual Range Air Combat Maneuvering Problem, Daniel B. Joseph

Theses and Dissertations

Artificial intelligence (AI) grows ever-more important in warfighting. Emerging technologies allow for the use of AI to control aircraft and weapons systems. This research investigates the application of reinforcement learning (RL) through the Proximal Policy Optimization (PPO) algorithm to a two-versus-two (2v2) beyond-visual-range (BVR) air combat maneuvering problem (ACMP). Implemented in the Advanced Framework for Simulation, Integration, and Modeling (AFSIM), the methodology frames the engagement as a Markov decision process, wherein an autonomous RL agent learns continuous control decisions—throttle, pitch, roll, and yaw—under a cooperative communication scheme. A multi-phase curriculum-learning approach facilitates the progressive acquisition of flight stability, weapon deployment, …