Digital Commons Network^™

Cybersecurity Center For Offshore Wind Energy (Final Project Round), Sachin Shetty

Center for Secure and Intelligent Critical Systems (CSICS) Publications

This project establishes a Cybersecurity Center for Offshore Wind Energy with the objective of designing and operating a cyber-physical testbed for wind energy farms (WEFs) that enables comprehensive cybersecurity research. The testbed incorporates a Supervisory Control and Data Acquisition (SCADA) system connected to turbine models via industrial-grade programmable logic controllers (PLCs) and remote terminal units (RTUs). It supports side-channel data acquisition, implementation and analysis of various cyberattack scenarios, and development of attack detection, mitigation, and best-practice guidance tailored to wind energy systems. During the project, the team expanded the number and fidelity of mathematical turbine models (MTMs), integrated these models …

Quantitative Assessment Of Roasted Coffee Freshness Over Time Using Multi-Parameter Analysis, Maya Fetzer

Honors Theses

Roasted coffee undergoes continuous chemical and physical degradation after roasting, yet industry freshness standards remain empirically defined rather than grounded in measurable parameters. This study developed a multi-parameter framework for quantifying roasted coffee freshness by integrating gas-phase and solid-phase analysis, backed by human sensory evaluation, across three roast levels — light (City+), medium (Full City), and dark (Vienna) — over a four-month storage period. CO2 off-gassing was monitored continuously for 30 day, gas chromatography-mass spectrometry (GC-MS) tracked volatile compound changes in headspace gas, and Fourier transform infrared-diffuse reflectance spectroscopy (FTIR-DRIFTS) quantified oxidation-associated functional group changes, and microindentation assessed bean stiffness. …

Optimizing Ucs Prediction Models Through Xai-Based Feature Selection In Soil Stabilization, Ahmed Mohammed Awad Mohammed, Omayma Husain, Mosab Hamdan, Abdalmomen Mohammed, Abdullah Ansari, Atef Badr, Abubakar Elsafi, Abubakr Siddig

Research Outputs: 2025-Present

Unconfined Compressive Strength (UCS) is a key parameter for the assessment of the stability and performance of stabilized soils, yet traditional laboratory testing is both time and resource intensive. In this study, an interpretable machine learning approach to UCS prediction is presented, pairing five models (Random Forest (RF), Gradient Boosting (GB), Extreme Gradient Boosting (XGB), CatBoost, and K-Nearest Neighbors (KNN)) with SHapley Additive exPlanations (SHAP) for enhanced interpretability and to guide feature removal. A complete dataset of 12 geotechnical and chemical parameters, i.e., Atterberg limits, compaction properties, stabilizer chemistry, dosage, curing time, was used to train and test the models. …

Analytics Framework For Cyberattack Risk, Severity, And Timing Prediction Using Survival - Aware Sequence Modeling, Mariia Voronina

Theses and Dissertations

This study aimed to develop an analytics framework for cyberattack risk, severity, and timing prediction using survival-aware sequence modeling. The primary objective was to analyze the temporal dynamics of cyberattacks by jointly considering attack type, severity, risk, and time-to-attack, and to evaluate whether these components can be effectively modeled within a probabilistic framework. The analysis was conducted using a synthetically generated cyberattack dataset designed for cybersecurity research purposes. The dataset included multiple attack categories, such as DDoS, Intrusion, Malware, severity levels, and engineered risk measures. Python was used as the primary programming language for data analysis and model development. The …

Applied Machine Learning In Mineral Prospectivity Mapping In The South Pass-Granite Mountains Of Wyoming, Carl D. York

Graduate Theses/Dissertations

The South Pass-Granite Mountains, comprised of Archean to Tertiary rocks embedded with an Archean greenstone belt in Central Wyoming, have historically been mined for gold, iron, steel, and silver. Rare earth elements have been of increasing interest in the region and thus, several datasets have become available. Knowledge driven and data driven models are used in mineral prospectivity mapping with input data including geological mapping, geochemical data, geophysical (USGS Critical Mineral magnetic and gravity data), radiometric (U, Th, K) data and multispectral remote sensing data. Mineral prospectivity has seen a rapid improvement in recent years using advancements in machine learning …

Machine-Learning- And Iot-Based Approach For Predicting Water Quality Using Data Classification And Explainable Ai Technique, M. Samir Abou El-Seoud, Omar H. Karam 2, Hosam El-Sofany 3

Computer Science

Maintaining water quality is crucial for both public health and environmental sustainability. This study proposes a machine-learning- (ML-) and Internet-of-Things- (IoT-) based approach to water quality prediction that utilizes explainable artificial intelligence (XAI) and data classification techniques. The proposed approach integrates IoT devices to enhance real-time data collection, facilitating continuous monitoring and early anomaly detection. Ten different ML classifiers—Decision Trees, K-Nearest Neighbors, XGBoost, Naïve Bayes, Logistic Regression, AdaBoost, Random Forests, Support Vector Machines, Voting, and Multi-Layer Perceptron—were evaluated to find out which one works best for predicting water quality. We used three distinct approaches for feature selection: analysis of variance …

Enlem: Ensemble Learning-Based Model To Detect Phishing Websites, Most Nilufa Yeasmin, Md Abu Rumman Refat, Bikash Chandra Singh, Zulfikar Alom, Zeyar Aung, Mohammad Azim

School of Cybersecurity Faculty Publications

Phishing involves manipulating individuals into revealing private data, e.g., user IDs, bank details, and passwords. The observed surge in fraud is related to increased deception, impersonation, and advanced online attacks. Thus, effective phishing detection methods are required to mitigate escalating global phishing threats. Existing methods (e.g., heuristics-based, signature-based, and visual similarity-based methods) attempt to detect phishing sites, and machine learning (ML) and deep learning (DL) methods are effective in the cybersecurity context in terms of learning from data, offering insights, and forecasting. However, independent ML algorithms are limited when handling complex data, and DL techniques surpass traditional ML methods in …

Machine Learning For Recession Prediction, Ethan Reusser

Williams Honors College, Honors Research Projects

Macroeconomic predictions present challenges in machine learning due to the rarity of economic recessions, the constantly-changing matter of global markets, and severe class imbalance in historical data. This project focuses on predicting the onset of United States economic recessions within a 12-month window using Python and Jupyter Notebook. A machine learning pipeline was developed utilizing multiple models: Logistic Regression, Random Forest, XGBoost, and Long Short-Term Memory (LSTM) neural networks. For class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was applied strictly to training data, paired with Platt scaling for calibration on thresholds. The resulting models were evaluated in the 2005 …

Machine Learning For Wearable Sensor-Based Human Movement Rehabilitation: A Five-Year Systematic Review, Yassine Benachour, Farid Flitti, Lina Maloukh, Aicha Beya Far, Elhocine Boutellaa, Mohamed Bentoumi, Marwa Chendeb El Rai, Nour Aburaed, Khaled Ali, Moez Rehman, Sultan Mosleh, Rania Dghaim, Sadok Bouamama

All Works

Wearable-sensor-based human movement analysis is an increasingly important component of digital health and rehabilitation, enabling objective monitoring and data-driven personalization of therapy. In parallel, machine learning (ML) methods have rapidly expanded for interpreting multimodal movement signals, yet the evidence base remains heterogeneous and difficult to benchmark. This PRISMA-guided systematic review synthesizes recent ML approaches for wearable human motion analysis in rehabilitation-oriented health applications. We searched IEEE Xplore, PubMed, and Scopus for English-language studies published from 2021 to 2025 and extracted information on sensor modalities, ML task formulations and model families, dataset characteristics, validation protocols, and reported performance metrics, together with …

A Bayesian Late-Fusion Supportability Framework For Rare-Disease Severity Prediction In Glut1 Deficiency Syndrome, Jordan M. Rodriguez

Mathematics Dissertations

Glucose transporter type 1 deficiency syndrome (GLUT1-DS) is a rare neurometabolic disorder with heterogeneous neurological and developmental severity. Because patient-level severity is not observed as a single validated outcome, this dissertation develops a Bayesian late-fusion supportability framework for constructing and predicting an ordered latent severity phenotype from clinical, genetic, and EEG-derived evidence. The primary target was constructed in a larger clinical cohort using age-5 symptom burden and learning cognition, then assigned to an aligned multimodal prediction cohort. Target-defining variables were excluded from supervised predictors, and models were evaluated using patient-exclusive cross-validation with training-fold preprocessing and fold-wise EEG PCA.

The primary …

Strain-Induced Nonvolatile Domain Switching And Tunable Elastic Modulus In Ba1-Xsrxtio3 Membrane By Phase-Field Simulation., Laveeza Ahmad

Material Science and Engineering Dissertations

Ferroelectrics underpin a broad spectrum of technological applications due to its switchable ferroelectric polarization and the associated electro-mechanical responses under electrical, optical, thermal, and mechanical stimuli. Recent advancement in membrane technology offers new opportunities to tune ferroelectric polarizations via mechanical strains at relatively large magnitude and scale. However, its influence on the tunability of mechanical responses of the membrane remains underexplored. Herein, we developed a phase-field model for free-standing Ba_1-xSr_xTiO₃ ferroelectric membranes with stress-free boundary conditions on top/bottom surfaces and achieved strain-induced nonvolatile ferroelectric domain switching in the membrane. It is discovered that a …

Modeling Sealed Deck Construction In Collectible Card Games Using Learning-To-Rank Approach, Michael Dinh Nguyen

Master's Projects

Artificial intelligence has demonstrated strong performance in complex decision-making domains such as chess and Go, motivating research into its application for games with even richer rules and combinatorial complexity. In collectible card games like Magic: The Gathering, deck construction from a constrained card pool is a critical and challenging task that requires evaluating card strength, synergy, and resource balance. This project explores whether machine learning, specifically learning-to-rank (LTR), can effectively model these human decision processes to construct competitive decks in a sealed format. The results found here can also be applicable to other areas of note, such as sports drafting …

Interpretable Machine Learning For In-Home Mild Cognitive Impairment Detection, Budhitama Subagdja, Shanthoshigaa D, Ah-Hwee Tan, Iris Rawtaer

Research Collection School Of Computing and Information Systems

This paper introduces a novel system for in-home cognitive health assessment using ambient sensors and a machine learning technology that can robustly detect mild cognitive impairment (MCI) despite limited available data. The learned model can explain the aspects of individuals’ daily lives led to the prediction, while reliably predicting MCI, providing more insights to healthcare workers for further clinical interventions. We developed the robust transparent machine learning model, based on fusion adaptive resonance theory (Fusion ART) neural network to learn individuals’ daily patterns of activity from continuous sensor data in terms of a suite of digital biomarkers reflecting four key …

Measuring The Efficiency Of The Arts And Culture Sector Relative To Gdp: A State-Level Analysis Of The United States, Ameerkhan Jaffar Khan Khader Khan

Selected Full-Text Master Theses 2021-

Arts and cultural production contributed $1.17 trillion to United States gross domestic product in 2023, 4.2% of the national total, but that contribution is spread very unevenly across states, and the official statistics describe how large the sector is rather than how efficiently it operates (Bureau of Economic Analysis, 2024). This study asks how efficiently each state convert growth in arts and culture employment into growth in value added, how many inputs the efficiency model can carry before it stops distinguishing 51 observations, and whether observable state characteristics explain the differences found. The analysis uses 2023 state-level data for all …

An Integrated Data-Driven Framework For Arctic Shipping: Analyzing Vessel Speed, Environmental And Ecological Factors Through Innovative Statistical Spatio-Temporal Methods, Inverse Optimization And Machine Learning, Mauli Pant

Theses and Dissertations

This dissertation develops an integrated data-driven framework to analyze vessel navigation and ecological risk in the United States Arctic from 2010 to 2019. As environmental change and maritime activity increase in the region, understanding how vessels respond to dynamic conditions and how those responses interact with marine ecosystems has become increasingly important. A central theme of this dissertation is the treatment of vessel speed as both an observed outcome and a decision variable reflecting trade- offs among operational, environmental, and ecological factors. The first chapter develops a predictive framework for vessel speed over ground (SOG) using Gaussian Process Boosting (GPBoost), …

Machine Learning-Based Intrusion Detection System For Iot Networks Using The Rt-Iot 2022 Dataset, Bukunmi Ebenezer Afolabi

Theses, Dissertations and Capstones

The rapid expansion of the Internet of Things (IoT) has transformed modern computing by enabling seamless connectivity among heterogeneous devices across diverse application domains. However, this increased interconnectivity has significantly enlarged the attack surface of IoT networks, exposing them to a wide range of sophisticated cyber threats. Conventional security mechanisms often lack the capability to detect emerging attacks in real time, thereby necessitating the development of intelligent Intrusion Detection Systems (IDS) capable of accurately identifying malicious network activities. This study developed and evaluated a machine learning-based intrusion detection framework for multiclass IoT attack detection using the RT-IoT2022 dataset. The dataset …

Machine Learning: Thematic Feature Grouping, And The Magnificent Seven: A Forecasting Analysis, Mirarmia Jalali, Mohammad Najand, Andrew Cohen

Finance Faculty Publications

This study examines the predictability of monthly excess returns for the “Magnificent Seven” U.S. technology firms using machine learning and economically motivated thematic feature grouping. Framed as a focused study of the most systemically consequential equity panel in modern markets—seven firms representing over 30% of the S&P 500—the analysis confronts a small-N, large-P environment where economically structured dimensionality reduction is essential. Using 154 firm-level characteristics categorized into 13 economic themes, we evaluate linear, penalized, tree-based, and neural network models in a small-N, large-P setting. Unrestricted models suffer substantial overfitting and fail to outperform the historical average benchmark out-of-sample. In contrast, …

Untrained Position-Encoded Multilayer Perceptron Network For Structured Illumination Microscopy Reconstruction, Sahil Sharma, Leonidas Zimianitis, Krishnendu Samanta, Balpreet Singh Ahluwalia, Joby Joseph, Dushan N. Wadduwage

Computer Science Faculty Publications

Structured Illumination Microscopy (SIM) enables super-resolution imaging by encoding high-frequency spatial information through patterned light. While traditional Fourier-based reconstruction methods are prone to artifacts under suboptimal conditions, recent deep learning approaches often require large training datasets and lack adaptability across different imaging setups. In this work, we present Position Encoded Multi-Layer Perceptron (PEM) network that leverages implicit neural representations (INRs) and SIM forward-model-driven modeling to reconstruct super-resolved images without any training data. PEM-SIM represents each spatial coordinate as a combination of sinusoidal functions across multiple frequencies, enabling rich encoding of fine spatial detail. A forward model grounded in SIM image …

Application Paths Of Semantic Modeling In Financial Fraud Detection And Risk Identification, Victor P. Gauthier, Daniel S. Wu

Computer Science Faculty Publications

Financial fraud and risk pose significant threats to economic stability and individual well-being. Traditional detection methods often struggle to keep pace with increasingly sophisticated fraudulent schemes. Semantic modeling, which focuses on understanding the meaning and relationships within data, offers a promising avenue for enhancing fraud detection and risk identification. This review paper explores the application paths of semantic modeling in this domain. We begin with a historical overview of fraud detection techniques, highlighting the limitations of traditional approaches. Subsequently, we delve into core themes, including knowledge graph-based fraud detection and semantic rule-based inference for risk assessment. We then compare and …

Assessment Of Carbon Capture Potential In Rubber Plantations Via Landsat 9 Imagery Analysis, Ade Fitria, Rahmawaty Rahmawaty, Razali Razali, Jamin Saputra, Seca Gandaseca

Baghdad Science Journal

Indonesia is a significant greenhouse gas (GHG) emitter, contributing 12.3% of carbon dioxide (CO₂) of total emissions. Carbon dioxide (CO₂), a major GHG, is increasing in the Earth’s atmosphere. The CO₂ absorption can be increased through rubber plantations because rubber plants, such as forest plants, can process CO₂ as a carbon source for photosynthesis. This research aims to analyze carbon uptake in relation to tree density, biomass of rubber vegetation, and soil organic carbon (SOC) content, as well as to map the distribution of carbon potential using remote sensing. This research was conducted in …

Automation And Autonomy In The Ivf Laboratory: Concepts And Implications For Embryologists, Jacques Cohen, Gerardo Mendizabal-Ruiz, Giles Anthony Palmer, Giuseppe Silvestri, Mina Alikani

EVMS School of Health Professions Faculty Publications

Automation is regarded as the next phase in the evolution of laboratory IVF. Despite technological advances, most laboratory procedures remain operator-dependent, contributing to variability in performance and outcomes. While automation has improved reproducibility and efficiency in many areas of medicine, its adoption in IVF has been slow. This review examines how automation may be integrated into IVF laboratories through a set of conceptual distinctions. Automation refers to the execution of procedural steps by machine-controlled systems, whereas autonomy describes the degree to which such systems can operate without human intervention. This review distinguishes between static automation, which maintains or monitors the …

Geospatial Analysis Of Sinkhole Density And Characteristics In West Virginia, Marissa Shannon Loftus

Graduate Theses, Dissertations, and Problem Reports (ETD)

Creation of a sinkhole map of West Virginia provides useful data to help identify areas susceptible to sinkhole related hazards, as well as allowing for detailed analysis of factors controlling karst development. Newly available, statewide LiDAR-derived digital elevation data have made mapping sinkholes over larger spatial extents feasible. For this study, sinkholes were mapped for all areas of West Virginia that are underlain by carbonate rocks; then the distribution and characteristics were evaluated overall and in four focus areas (Berkeley and Jefferson counties, Pendleton County, Greenbrier County, and Monroe County). These areas correspond to broader physiographic regions, with Berkeley, Jefferson, …

Artificial Intelligence And Machine Learning In Smart Vaginal Formulation Development, Deborah A. Ogundemuren, Vivek Agrahari, Andrew P. Wong, Carolina Herrera, Margaret O. Ilomuanya, Gustavo F. Doncel

CONRAD Publications

Vaginal drug delivery in women's health remains underutilized and insufficiently studied, largely due to the complexity and dynamic nature of the vaginal microenvironment. Variations in vaginal pH, hormonal levels, and microbiota composition introduce significant biological variability, complicating formulation design and contributing to inconsistent therapeutic outcomes and poor patient adherence. Conventional vaginal formulations often fail to account for these individual differences, highlighting the need for more adaptive and predictive approaches. Emerging advances in artificial intelligence (AI) and machine learning (ML) offer promising strategies to address these challenges by enabling multi-parameter, data-driven formulation development that explicitly considers biological variability. Despite their transformative …

The Crucial Role Of Machine Learning Models In Predicting Current Childhood Asthma: Model Comparison, Calibration, And Shap-Based Interpretation, Aditya Chakraborty, A. K.M. Raquibul Bashar

Epidemiology, Biostatistics, & Environmental Health Faculty Publications

Background: Asthma is one of the most prominent chronic diseases in children and one of the most challenging ailments to diagnose in infants and preschoolers in the United States. Predictive models can be instrumental in improving early diagnosis, personalized treatment strategies, and disease progression. By utilizing nationalized data, this study focuses on building and comparing high-performing analytical predictive models based on the relevant risk factors and identifying the most influential predictors.

Methods: We analyzed cross-sectional BRFSS Asthma Call-Back Survey data (2011-2020; N = 9,813) and randomly split participants into training and testing sets. An XGBoost model (hyperparameters tuned via grid …

Pd-L1-Centric Whole Blood-Based Immune Signature Profiles Of Tuberculosis Patients During Therapy, Johanna Eggeling, Martina Sester, Christoph Lange, Jan Heyckendorf, Barbara Kalsdorf, Anna M Mandalakas, Andrew R Dinardo, David Lewinsohn, Dagmar Schaub, Tina Schmidt, Eva Tolosa, Maja Reimann, Patricia M Sánchez Carballo

Faculty, Staff and Students Publications

This prospective study investigated whole blood-based immune cell biomarkers for pulmonary tuberculosis (TB) immunoprofiling. Blood samples from 34 healthy controls and 51 tuberculosis patients were analyzed at three timepoints: Prior to therapy (T0), after 14 days of therapy (T1), and at the end of treatment (Te). Using multiparameter flow cytometry, 386 immune cell populations were analyzed. Predictive models were developed using two machine learning algorithms. A TB5LF change score, which was based on five cell populations, effectively distinguished tuberculosis patients from controls (AUC = 0.89) and tuberculosis patients before and at the end of treatment (AUC = 0.92). Similarly, the …

Saudi Stock Market Prediction Using Majority Voting, Mobeen W. Alhalabi, Aiiad Albishri

Journal of King Abdulaziz University: Computing and Information Technology Sciences

Predicting the stock market via data analysis is an important research topic. Because the system implementing it seems to think and act like humans, for the last two decades’ machine learning got a lot of attention from developers, programmers, scientists, even from the general public. Due to their advantages, machine learning techniques can be used in many areas, including finance and stock markets. Although many studies were performed with excellent results for the investors, many researchers are still looking for better results that will help minimize risk with higher percentage. In this study, the state of the techniques to predict …

Data-Driven Analysis And Atomistic Simulations Of Next-Generation Materials For Energy Conversion And Storage, Yuliang Shi

Dissertations

Metal-organic frameworks (MOFs), with their modular architectures and tunable properties, represent an especially rich domain for accelerated material design and discovery for a range of diverse applications. Within this class of multifunctional materials, two-dimensional (2D) electrically conductive MOFs (EC MOFs) are of particular interest, as their 7r-stacked layered structures combine permanent porosity with electronic conductivity, enabling potential breakthroughs in energy storage, energy conversion, and quantum sensing. But the discovery and design of new EC MOFs based on expensive experimental screening is increasingly impractical due to the infinite chemical space. Furthermore, the practical implementation of EC MOFs for specific tasks depends …

Applying Machine Learning Techniques For Early Detection Of Cyber Attacks On Iot Devices, Noor Adnan Allamy

Al-Esraa University College Journal for Engineering Sciences

This research designs, implements, and evaluates a machine learning-based framework for the early detection of cyber attacks targeting Internet of Things (IoT) devices, with a specific focus on the context and challenges present in Iraq. The study conducts a comparative analysis of three supervised learning algorithms—Support Vector Machine (SVM), Random Forest (RF), and Deep Neural Networks (DNN)—using a combination of benchmark datasets (NSL-KDD, CIC-IDS-2017, Bot-IoT) and a synthesized dataset adapted to simulate the Iraqi threat landscape. Key performance metrics, including accuracy, precision, recall, and F1-score, were used for evaluation. The proposed Random Forest model demonstrated superior performance, achieving an accuracy …

Deep Learning Approaches For License Plate Detection In Saudi Arabia, Reemah M. Alhebshi, Abeer Almakky, Lujain Waleed Aljahdali, Raghad Bakr Hawsawi

Journal of King Abdulaziz University: Computing and Information Technology Sciences

The most important and challenging tasks in an intelligent traffic monitoring system are vehicle detection and classification. Conventional approaches are significantly computationally intensive and create limitations when the data-collecting modality alters. This study evaluates the effectiveness of deep learning models in extracting and classifying key vehicle attributes, such as license plates, models, and colors, through four AI-driven components for vehicle analysis. The YOLOv8m model was used for detecting the vehicle license plate and character recognition, achieving accuracy ranging from 95.10% to 99.24%. For vehicle model recognition, the Xception model also showed high performance with a precision of 96.79%. The EfficientNetB3 …

Towards Computational Methods In Medical Data Analysis: From Speech And Text To Imaging, Kristin Qi

Graduate Doctoral Dissertations

Early detection of cognitive decline and efficient medical image analysis remain critical challenges in healthcare. Traditional clinical assessments are infrequent and resource-intensive, while everyday speech data and unlabeled medical images remain largely unexploited. This dissertation develops computational methods integrating machine learning and artificial intelligence across speech, text, and imaging modalities to address challenges in medical data processing. For cognitive monitoring, this work first introduces methods using voice assistant systems to collect longitudinal speech data in home environments, demonstrating that incorporating historical session patterns significantly enhances detection of mild cognitive impairment. Building on this foundation, a framework combining large language model-driven …