Digital Commons Network^™

Artificial Intelligence And Translation: Exploring Current Applications, Limitations And Future Potential Of Language Models Through Japanese-English Translation, Loklin Elias Nord

Undergraduate Theses, Capstones, and Recitals

This thesis highlights the recent improvements and capabilities of Large Language Models (LLMs), specifically their ability to produce translations between different languages. The continued up-scaling of model sizes has led to breakthroughs in the level of their observed intelligence, allowing them to produce translations that are similar in quality to highly skilled human translators. However, to facilitate the reasoning processes that LLMs now possess, their demand for computational power and the supporting hardware and resources has increased proportionally. Considering the impacts of this technology on the environment, energy resources, and its accessibility, my research explores the possibilities of smaller, highly …

Ai Failures In The Eyes Of The Downstream Developer: A First Look At Concerns, Practices, And Challenges, Haoyu Gao, Mansooreh Zahedi, Wenxin Jiang, Hong Yi Lin, James C. Davis, Christoph Treude

Research Collection School Of Computing and Information Systems

With the advancement of AI models, more software systems are adopting AI as a component to facilitate automation. Pre-trained models (PTMs) have become a cornerstone of AI-based software, allowing for rapid integration and development with lower training cost. However, their adoption also introduces failure modes such as data leakage and biased outputs, that may require careful handling by downstream developers. While previous research has proposed taxonomies of these technical concerns and various mitigation strategies, how downstream developers address these issues during the development of general AI-based software when reusing PTMs remains unexplored. Understanding downstream developers’ perspectives is essential because they …

Bridging The Gap Between Career Expectations Versus Labor Market Realities, Reinette P. Madrid, Grethel T. Ledesma, Ignatius Aryono Putranto, Jyro B. Triviño

Leadership and Strategy Faculty Publications

Most students lack awareness regarding the labor market outcomes for their chosen college majors. This study aims to answer what factors affect the career expectations of graduating students at Quezon City University and how these expectations align with the prevailing labor market situation. It employed descriptive, causal, and explanatory research using a sample of 108 respondents from fourth-year information technology students for the school year 2021 to 2022. Eight of the nine null hypotheses were rejected by employing multinomial logistic and linear regression. Student fixed effects and other labor market outcomes significantly predicted salary, estimated stability, and estimated skills in …

Logupdater: Automated Detection And Repair Of Specific Defects In Logging Statements, Renyi Zhong, Yichen Li, Jinxi Kuang, Wenwei Gu, Yintong Huo, Michael R. Lyu

Research Collection School Of Computing and Information Systems

Developers write logging statements to monitor software runtime behaviors and system state. However, poorly constructed or misleading log messages can inadvertently obfuscate actual program execution patterns, thereby impeding effective software maintenance. Existing research on analyzing issues within logging statements is limited, primarily focusing on detecting a singular type of defect and relying on manual intervention for fixes rather than automated solutions.To address the limitation, we initiate a systematic study that pinpoints four specific types of defects in logging statements (i.e., statement code inconsistency, static dynamic inconsistency, temporal relation inconsistency, and readability issues) through the analysis of real-world log-centric changes. We …

Generalizing Unsupervised Lidar Motion Estimation Via Patch-Aware Feature Reweighting Under Snowy Conditions, Beibei Zhou, Zhiyuan Zhang, Zhenbo Song, Jianhui Guo, Tongpo Zhang, Shaojing Song, Qinge Wu, Hui Kong

Research Collection School Of Computing and Information Systems

Unsupervised LiDAR motion estimation has attracted significant attention in pattern recognition and robotic vision. However, ensuring robust performance under adverse conditions — particularly structured noise like heavy snowfall — remains a challenging problem. Existing models struggle with domain generalization due to the sensitivity of feature extractors to spatial outliers, limiting their reliability in complex real-world scenarios. In this work, we present an unsupervised LiDAR motion estimation framework via patch-aware feature reweighting to mitigate the domain shift between clear and snowy conditions. Our approach prioritizes reliability-aware reweighting to reduce the influence of snowflake noise and outlier points on pose estimation, while …

Push-Pull Farming System Helps Insure Smallholder Maize Production Under Climate Change, Yann Clough, Wilhelm May, Timothy Luttermoser, Dickens Nyagol, Nikolaos Alexandridis, Frank Chidawanyika, Mattias Jonsson, Charles A.O. Midega, Katja Poveda, Zeyaur R. Khan

All Peer-Reviewed Publications

Agroecological approaches harnessing in-field biodiversity have been highlighted as a sustainable way to reduce yield gaps, but long-term assessments of their ability to buffer production from weather extremes and increasing interannual and intraseasonal climate variability are scarce. Here, we analyze a 12-year dataset from 642 control vs. push-pull maize field pairs in western Kenya, where long-term climate trends include increasing temperatures and intensifying drought and rainfall. In push-pull, maize is intercropped with a fodder legume that improves soil fertility, suppresses weeds, and deters pests, and is bordered with fodder grasses attracting and suppressing the pests. We assess interactions between push-pull …

An Assessment Of Carbon Dioxide Removal Potential From Stormwater Ponds Using Concrete-Based Materials, Yu-Hsuan Tai

Theses and Dissertations (Comprehensive)

Stormwater ponds (SWPs) are engineered stormwater infrastructure known to be major greenhouse gas (GHG) emitters, including carbon dioxide (CO₂). Concrete-based materials have great potential to capture CO₂ as carbonate minerals (CaCO₃), through aqueous carbonation, driven by their alkaline nature and high portlandite (Ca(OH)₂) content. However, the mechanistic understanding of how specific parameters including mass, surface area, and degradation of calcium silicate hydrate (C-S-H) phases collectively control carbonation kinetics and CO₂ uptake under dynamically evolving conditions remains underexplored. Laboratory-scale experiments were conducted to evaluate the effects of cement dosage, particle surface area, and …

Time Series Mediation Analysis With Non-Linear And Machine Learning Methods, Timothy Marsh

Theses and Dissertations (Comprehensive)

Non-linearity in mediation analysis has been primarily studied in the context of binary variables which encode a treatment and control, estimating a `direct` and `indirect` effect of a covariate (a.k.a. treatment) X on a response Y, with a third variable M (the `mediator`) that is affected by X and in turn affects Y. The focus of mediation analysis in general is to quantify the overall effect of X on Y, including the effect through M. This presentation will focus on one or more continuous treatments and apply non-linear methods, including splines and machine learning models, to mediation analysis in a …

Likelihood-Based Inference For The Poisson–Tweedie Family, Si Chen

Theses and Dissertations (Comprehensive)

This thesis develops likelihood-based methodology for flexible modeling and inference for heterogeneous count data within and adjacent to the Poisson–Tweedie family. A mean–dispersion regression framework is introduced to jointly relate the mean and dispersion to covariates, enabling separation of covariate-driven heterogeneity from distributional parameters. Likelihood inference is implemented using recursive probability mass function evaluation, and performance is assessed through simulation and applications including RNA-sequence counts. The framework is extended to accommodate excess zeros via zero-inflated Poisson–Tweedie models, with full likelihood-based estimation and improved risk estimation illustrated using insurance claim data. Finally, boundary behaviour is investigated through the discrete stable distribution …

Toward Mapping Multiphase Multicomponent Mixtures With Neural Networks, Kristen L. Hallas, Melissa De Jesus, Christine J. Wu, Jianzhi Li, Jason Bernstein, Philip C. Myint

School of Mathematical & Statistical Sciences Faculty Publications

Equation of state (EOS) tables are commonly used in hydrodynamic simulations of high-pressure, high-temperature phenomena in fields like planetary science, astrophysics, and high-energy-density science. However, generating and storing EOS tables for multiphase, multicomponent mixtures over a wide range of pressures and temperatures is computationally infeasible due to their memory-intensive nature. To address this issue, we have developed a neural network-based machine learning model to predict new EOS tables for binary mixtures. In particular, a deep feedforward neural network trained on a set of ten EOS tables at particular mixture compositions is able to predict nine new (hold-out) EOS tables at …

Enhancing Shipboard Safety Management Under The Ism Code: An Innovative Risk Assessment Framework With A Stern Tube Case Study, Pi-Yen Lin

Journal of Marine Science and Technology–Taiwan

The shipboard safety management system (SMS) is designed to enhance safe operations, risk management, and emergency response to improve overall ship safety and efficiency. This paper demonstrates the use of an engine room simulator (ERS) for collecting failure modes and applies it to a comprehensive failure analysis of the stern tube lubricating oil system. A new risk closeness coefficient method was developed, integrating expert background knowledge and weighted risk assessments. The analysis, based on multiple expert evaluations, covered five subsystems, eight main components, 23 failure modes, and 112 failure causes. This study presents 26 recommendations for maritime practitioners and onboard …

Robust Statistical Methods For Microbiome Abundance Data, Yiming Shi

WUSM Theses and Dissertations – All Programs

Differential abundance analysis in microbiome studies aims to identify taxa whose abundance differs across biological or clinical conditions. The observed data are typically taxon-specific sequencing read counts, representing reads assigned to different taxa within each sample. These counts are indirect measurements of the underlying microbial abundance profile and are constrained by sample-specific library sizes. Microbiome count data are also typically sparse, overdispersed, and heteroscedastic. Together, these characteristics create substantial challenges for differential abundance analysis and make the results highly sensitive to normalization procedures, model specification, and the statistical methods used for inference.

Normalization defines the scale on which samples are …

Port And Vessel Communication Traffic Intrusion Detection: A Variational Autoencoder‑Enhanced Multilayer Perceptron Approach, Chien-Lin Chiang, Hsien-Cheng Chou, Ming-Yuan Peng, Yi-Yuan Chiang, Yu-Shun Liu

Journal of Marine Science and Technology–Taiwan

Port and vessel networks increasingly operate on IP/Ethernet backbones with high‑noise, high‑dimensional traffic. We present a lightweight hybrid intrusion‑detection model that couples a variational autoencoder (VAE) with a multilayer perceptron (MLP) and augments training with a boundary‑oriented latent‑space mixup strategy. The VAE models the distribution of normal traffic and identifies anomalies through reconstruction errors. Subsequently, it generates robust latent vectors, enabling the MLP to perform highly accurate supervised classification. On the UNSW‑NB15 dataset, the proposed pipeline attains ≥97% accuracy and an outstanding recall of 99.56% in binary intrusion detection, and visualization of the latent space (PCA) together with reconstruction‑error analyses …

Sar Ship Detection Based On Shallow Feature Guidance, Chenxu Xia, Peng Chen, Ya Zhang, Ying Li

Journal of Marine Science and Technology–Taiwan

Maritime ship detection is of great significance for both military security and civilian applications. Synthetic Aperture Radar (SAR), with its all-weather and all-day imaging capability, plays a vital role in maritime surveillance. Nevertheless, SAR ship targets typically appear small in scale, embedded in complex backgrounds, blurred at boundaries, and easily confused with near-shore features, which pose substantial challenges for accurate detection. To address these issues, we propose a SAR ship detection network that integrates dual enhancements of small-object representation and edge information. The network introduces two key components: the Small Target Refine Pyramid (STRP) to strengthen shallow feature representation for …

Hed-Id: A Federated Expert System For Interpretable And Resource-Adaptive Intrusion Detection On Edge Devices, Kushboo Nasir, Sahar K. Badri, Asad Khattak, Daniyal M. Alghazzawi, Mohammed Yahya Alghamdi, Mona Alkhozae, Abeer Almakky, Rania M. Alhazmi, Muhammad Zubair Asghar

All Works

The proliferation of edge computing in industrial and IoT networks necessitates expert systems that support accurate, interpretable, and resource-efficient intrusion detection under strict privacy and computational constraints. This paper presents HED-ID, a fully integrated federated expert system designed for real-time, edge-oriented anomaly detection. The system combines: (i) a stacked bidirectional GRU (BiGRU) architecture with hybrid temporal–spatial attention for sequential and feature-level anomaly modeling, (ii) a resource-adaptive Grey Wolf Optimizer (A-GWO) that modulates exploration–exploitation behavior based on real-time CPU utilization and Bayesian epistemic uncertainty, with formal convergence proof under bounded resource shifts (Appendix A), (iii) federated SHAP, which aggregates local explanations …

Fine-Tuning Large Language Models For Binary And Multiclass Target-Specific Hate Speech Detection, Sanaa Kaddoura, Sumaia Al-Kohlani

All Works

Social media platforms are widely used by individuals from diverse demographic backgrounds to share their opinions and daily experiences. The anonymity provided by these platforms has increased the prevalence of hate speech, causing significant psychological harm to targeted groups. Existing hate speech detection approaches predominantly rely on binary classification, which fails to capture the lexical and contextual diversity of hate categories, leading to the misclassification of hate against minority groups. This study addresses this limitation by evaluating the effectiveness of five large language models (LLMs) for target-specific multiclass hate speech detection across six classes. The evaluation uses a dataset of …

Enhancing Agricultural Sustainability Under Climate Change: A Multi-Scale Framework Integrating Climate Extremes, Resource Efficiency, And Data-Driven Modeling, Shahryar Fazli

Computational and Data Sciences (PhD) Dissertations

Agricultural systems are increasingly challenged by climate variability, where shifting temperature regimes, hydrological variability, and the rising frequency of compound and cascading extremes threaten global food security and resource sustainability. Addressing these challenges requires integrated frameworks that bridge biophysical monitoring, predictive modeling, and adaptive decision-making. This dissertation develops a data-driven, multi-scale framework to quantify and enhance agricultural resilience by integrating remote sensing, climate analytics, and machine learning across the United States, with a focus on California and the Western U.S.

First, hyperspectral and thermal remote sensing data from EMIT and OpenET are integrated to characterize crop nitrogen–water interactions and assess …

Revisiting Ulam Stability For Boundary Value Problems, Martin Bohner, Snezhana Hristova, Agnieszka B. Malinowska, Ewa Girejko

Mathematics and Statistics Faculty Research & Creative Works

The main goal of this paper is to apply Ulam stability theory to boundary value problems for dynamic equations, while addressing several common misconceptions found in the existing literature. We identify the key issues that arise when applying Ulam stability to such problems and propose three distinct approaches to overcome them. To enhance clarity and accessibility, we begin with nonlinear ordinary differential equations and subsequently extend the analysis to nonlinear dynamic equations on time scales. Since a time scale is defined as any nonempty closed subset of the real numbers, our results are applicable to dynamic equations on continuous, discrete, …

Wildfire And Fuel Treatments In Utah: A Practical Guide And Outcomes From An Oak-Maple Ecosystem, Annie E. Prescott

All Graduate Theses and Dissertations, Fall 2023 to Present

Wildfire is a natural part of Utah’s landscapes, but as wildfires become more extreme and communities continue to expand into wildland areas, it is increasingly important to understand how wildfires behave and how their negative effects can be mitigated. One way to reduce wildfire risks and protect life, property, and ecosystem services is through fuel treatments — actions that reduce the amount and arrangement of flammable vegetation through mechanical removal, wildland fire, or other methods.

Information on wildfire behavior and fuel treatments is abundant but often scattered, ecosystem-specific and difficult to access. In addition, most research in the Western U.S. …

Toward Optimizing Fruit Production In Utah Soils: Apple Salt Tolerance Screening And Tart Cherry Fertility Management, Kaitlin Dabbs

All Graduate Theses and Dissertations, Fall 2023 to Present

Fruit production in Utah can be challenging due to local climate and soil conditions. Tart cherries and apples are major fruit crops in Utah, and improvements to these cropping systems would be beneficial for farmers and local consumers. Two experiments were conducted to evaluate desired improvements for these crops. The first study looked for salt tolerance in ten new apple rootstocks. Salty soils can be found throughout the state limiting the land suitable for apple production. When grown in salty soils apple trees often have reduced growth and production potential. A specialized drip irrigation system was constructed in a greenhouse …

A Projector-Rank Partition Theorem For Exact Degrees Of Freedom In Experimental Design, K. G. Nagananda

Mathematics and Statistics Faculty Publications and Presentations

In many experimental designs—split-plots, blocked or nested layouts, fractional factorials, and studies with missing or unequal replication—standard ANOVA procedures no longer tell us exactly how many independent pieces of information each effect truly contributes. We provide a general degrees of freedom (df) partition theorem that resolves this ambiguity. For N observations, we show that the total information in the data (i.e., N −1 df) can be split exactly across experimental effects and randomization strata by projecting the data onto each stratum and counting the df each effect contributes there. This yields integer df—not approximations—for any mix of fixed and random …

Quantifying Customer Sentiment For Automobile Brand Perception Analysis Using Machine Learning On Twitter, Sujith Samuel Mathew, Kadhim Hayawi, Neethu Venugopal, May El Barachi

All Works

Social networking sites provide a platform for individuals to express their opinions publicly. Brand managers actively use these platforms to gain insights into brand perceptions, as users often share their views on products and services. In this study, we use sentiment analysis to assess customer sentiment towards five leading automobile brands, analyzing text content shared on Twitter(or X). The research models the ’Brand Polarity Score’, which indicates whether customers perceive the brand positively or negatively. This score is further weighted based on the tweet’s influence, characterized by the engagement metrics of the tweet and the author’s follower count. We also …

Dblp: Noise Bridge Consistency Distillation For Efficient And Reliable Adversarial Purification, Chihan Huang, Belal Alsinglawi, Islam Al-Qudah

All Works

Recent advances in deep neural networks (DNNs) have led to remarkable success across a wide range of tasks. However, their susceptibility to adversarial perturbations remains a critical vulnerability. Existing diffusion-based adversarial purification methods often require intensive iterative denoising, severely limiting their practical deployment. In this paper, we propose Diffusion Bridge Distillation for Purification (DBLP), a novel and efficient diffusion-based framework for adversarial purification. Central to our approach is a new objective, noise bridge distillation, which constructs a principled alignment between the adversarial noise distribution and the clean data distribution within a latent consistency model (LCM). To further enhance semantic fidelity, …

X-Threat Framework For Adaptive And Explainable Deep Learning-Based Minority-Class And Zero-Day Cyber Threat Detection, Samina Naz Qaisarani, Sahar K. Badri, Asad Khattak, Daniyal M. Alghazzawi, Mohammed Yahya Alghamdi, Mona Alkhozae, Abeer Almakky, Rania M. Alhazmi, Muhammad Zubair Asghar

All Works

The dynamic nature of cyber threats—particularly minority-class and zero-day attacks—poses significant challenges to existing intrusion detection systems (IDS), which often lack adaptability, interpretability, and robustness. This paper presents X-THREAT, an adaptive and explainable deep learning framework designed to improve detection of rare and previously unseen cyber threats. Unlike prior approaches that treat detection, augmentation, and explainability separately, X-THREAT integrates these components within a unified pipeline. A hybrid VAE–GAN module enhances minority-class representation during training, while a CNN–BiLSTM–attention model performs threat detection, and a context-aware SHAP mechanism provides interpretable insights into model decisions. Experimental results on three benchmark datasets—CICIDS2017, UNSW-NB15, and …

Quasi-Poisson Analogs Of Regular Poisson Varieties, Casen Thompson

All Graduate Theses and Dissertations, Fall 2023 to Present

Let G be a complex semisimple linear algebraic group with 𝐶 a conjugacy class of parabolic subgroups of G. In previous work, Dr. Crooks defines 𝑢_𝐶→ B_𝐶 , a universal flat family of a fine Hamiltonian Lagrangian G-bundles over 𝐶, and shows that 𝑢_𝐶 is a family of regular Poisson varieties. This manuscript introduces a natural candidate for a quasi-Poisson analog of Dr. Crooks’s construction, 𝑢_𝐶, using the framework of quasi-Poisson geometry established by Alekseev, Kosmann-Scwarzbach, and Meinrenken. It will be shown that this proposed quasi-Poisson analog indeed has a valid quasi-Poisson structure and …

Reduced Mediators Released By Cyanobacteria During Exoelectrogenesis Detected Using Differential Pulse Voltammetry, Laura T. Wey, Monica Brachi, Yagut Allahverdiyeva, Shelley D. Minteer

Chemistry Faculty Research & Creative Works

Cyanobacteria generate electrical current through exoelectrogenesis (extracellular electron transfer) downstream of photosynthesis, yet the identity of the endogenous redox mediator(s) responsible remains unresolved. Here, we apply differential pulse voltammetry (DPV) to detect and characterise redox-active species released by Synechocystis sp. PCC 6803 under illumination. To enable sensitive detection, we developed an electrolyte intermediate between BG11 (Blue-Green-11) growth medium and MOPS (3-(N-morpholino)propanesulfonic acid) buffer that minimises background electrochemical interference while maintaining short-term cellular functionality. DPV revealed multiple light-enhanced oxidation peaks (0.1–0.65 V vs. saturated calomel electrode (SCE)) that were not resolvable using cyclic voltammetry, providing evidence that cyanobacteria release reduced compounds …

Generative Ai Adoption And Solvers' Popularity On Supply-Driven Crowdsourcing Platforms: The Dual Role Of Price Signals, Zimeng Zhu, Carol Hsu, Fiona Fui-Hoon Nah, Na Liu

Research Collection School Of Computing and Information Systems

Purpose – We investigate the effect of solvers’ adoption of Generative AI (GenAI) on their popularity in a supply-driven crowdsourcing platform. We also examine the impact of price signals as well as their heterogeneous impact based on the solvers’ membership duration on the platform. Design/methodology/approach – Our analysis focuses on solvers who adopt GenAI for design-related gigs on the supply-driven crowdsourcing platform. By combining propensity score matching (PSM) with multi-period difference-in-differences (DID), we examine how GenAI adoption impacts solvers’ popularity and how price signals affect this main effect. Findings – Our findings reveal that solvers who adopt GenAI tend to …

Artificial Intelligence (Ai) For Social Innovation In Health Education: Promoting Health Literacy Through Personalized Ai-Driven Learning Tools – A Systematic Review, Dina Mansour Tbaishat, Maha Waleed Elfadel

All Works

Background: Artificial Intelligence (AI) is transforming health education by enabling personalized, adaptive, and scalable approaches that may enhance aspects of health literacy. Despite rapid adoption, comprehensive synthesis of AI tools’ impact on health literacy as social innovation is limited. Understanding these effects guides educators, developers, and policymakers in designing potentially effective, inclusive, and ethical AI interventions. This review examines generative AI models, chatbots, and adaptive learning systems in supporting health literacy globally. Methods: A systematic review was conducted following PRISMA guidelines. Literature was identified primarily through PubMed/Medline, Scopus, and ScienceDirect. Connectedpapers.com was used exclusively as a citation chasing tool, performing …

Radio Frequency Tagging–Enabled Patient Monitoring: Integrating Mobility Tracking With Early Warning Systems For Enhanced Safety, Ahed Abugabah, Prashant Kumar Shukla, Piyush Kumar Shukla, Abhishek Dwivedi

All Works

Ensuring patient safety in healthcare environments requires continuous monitoring systems capable of identifying early warning signs of clinical risk. Traditional surveillance methods often fail to capture meaningful patterns in patient movement, limiting their ability to prevent incidents such as falls, prolonged immobility, or unnoticed health deterioration. Radio Frequency Tagging technology has been increasingly adopted for real-time patient tracking; however, existing systems are generally limited to location detection and lack predictive insights into patient behaviour. To overcome these limitations, this study presents a Radio Frequency Tagging-based patient monitoring framework that integrates mobility tracking with an early warning mechanism to enable proactive …

Understanding Chatbot-Assisted Collaborative Learning Among Female Undergraduate Students, Mohammad Amin Kuhail, Ahmed Shuhaiber, Sinan Salman, Nazik Alturki

All Works

Computer programming can be daunting for beginners due to complex concepts and syntax. Traditional teaching methods, while engaging through gamification and active learning, often lack personalized approaches. Recent advancements in artificial intelligence (AI), particularly large language models (LLMs), present new possibilities for personalized and interactive learning environments. This study introduces a chatbot-assisted collaborative learning environment (CCLE) that leverages an LLM (GPT-4) to enhance collaborative programming education. The CCLE enables real-time guidance and collaboration through natural language interactions, allowing students to work together on programming tasks, edit code collaboratively, and engage with both peers and the educational chatbot. We conducted an …