Digital Commons Network^™

Language In The Teaching Or: Expressing Confidence Versus Community, Aarya Ramprasad, Imaima Casubhoy, Melanie Meister, Brenda Bethman, Austin Bachar, Gary Sutkin

Gulf Coast Division GME Research Day 2025

No abstract provided.

Workarounds In American Public Law, Daniel A. Farber, Jonathan S. Gould, Matthew C. Stephenson

Faculty Articles

A workaround is a maneuver that seems, on its face, consistent with the formal rules, but that employs those rules in an unanticipated way to circumvent a legal obstacle. Though some workarounds are tolerated or even celebrated, workarounds (and proposed workarounds) often provoke instinctive skepticism or hostility. When, if ever, is such skepticism justified? Do workarounds raise distinctive legal or public policy concerns? This Article seeks to provide a systematic normative assessment of workarounds in American public law. We argue, first, that from a general public interest perspective, the desirability of a workaround depends primarily on the desirability of the …

Globalization In Context, Ayu Okvitawanli, Wolfgang Friedlmeier, Rachana Bhangaokar

IACCP Proceedings of the Biennial International Conferences

The Proceedings are divided into two parts: Volume I includes the links to the recorded keynotes and the book of abstracts of the XXVII Congress of the International Association for Cross-Cultural Psychology, 2024. Volume II consists of the individual empirical papers.

The Political Question Of Public Lands, Ryan A. Semerad

Wyoming Law Review

The Property Clause of Article IV of the United States Constitution provides: “Congress shall have Power to dispose of and make all needful Rules and Regulations respecting the Territory or other Property belonging to the United States.” Thus, among other things, the Constitution assigns the disposal of federally owned lands to Congress alone. That means the other branches of the federal government, including the courts, have no jurisdiction to exercise this power or to compel Congress to use it. Put another way, when Congress does not dispose of public lands through sale or transfer, that inaction is not subject to …

Days Of Future Past: The Legacy And Self-Efficacy Of Black And Hispanic Educators In Predominantly White School Districts, Tesfa S. Stewart

Selected Full Text Dissertations, 2011-

This mixed-methods dissertation investigates the self-efficacy of Black, Hispanic, and Multiracial educators in predominantly White institutions (PWIs), examining its connections to job satisfaction, career development, and perceived impact on students. Data from 34 participants, collected using the Teachers’ Sense of Efficacy Scale (TSES), indicated high overall self-efficacy, particularly in instructional strategies and classroom management. A one-way ANOVA revealed a statistically significant effect of race/ethnicity on total self-efficacy scores, with Black educators reporting significantly higher scores than their Hispanic and Multiracial peers (η² = .271). A simple linear regression using years of educational experience as the sole predictor did not produce …

How Self-Perceptions Relate To Recorded Proficiency: An Action Research Project Aimed At Implementing A Supplemental Reading Program For Balinese Students, Katherine L. Johnson

2025 Fall Honors Capstones Projects - Archive

English as a Foreign Language (EFL) education has had a complex history globally. This is especially true for Bali, Indonesia, where tourism is the main driver of the local economy. Learning English is highly valued, yet many students are left struggling to build the skills needed to rise above the local poverty line. In this action research study, the primary goal was to analyze students’ reading proficiencies to recommend the implementation of a reading program that could be appropriately tailored to individual students. This was done through administering a reading test, tracking both comprehension and reading speed, and a survey …

Hybrid Agentic System For Schema-Aware Nl2sql Generation, David Omondi Onyango

All Graduate Theses, Dissertations, and Other Capstone Projects

The natural language to SQL (NL2SQL) task enables non-expert users to interact with relational databases via natural language interfaces. However, NL2SQL frameworks often rely on Large Language Models (LLMs), raising concerns about computational overhead, data privacy, and deployment in resource-limited environments. To address these issues, we propose a hybrid schema-aware agentic system using Small Language Models (SLMs) as primary agents, with a selective LLM fallback mechanism. The LLM activates only when errors are detected in SLM-generated queries, reducing inference costs. Experiments on the BIRD benchmark dataset show our system achieves an execution accuracy of 53.91% and validation efficiency score of …

Understanding The Intersection Between Alaska Native And Military Cultural Identities: Using Arts-Based Narrative Inquiry With Mixed Methods, Leslie M. Robertson

Art Therapy | Doctoral Dissertations

American Indian and Alaska Native (AI/AN) peoples have contributed significant efforts towards involvement in the military and have served in the Armed Forces at five times the national average. AI/AN peoples have served with distinction in every major conflict for over 200 years. There has been a pervasive and historical lack of recognition for their service and a lack of awareness of the intersection between Indigenous and military cultures (Ault, 2020; Bennett, 2009; Department of Veterans Affairs, 2020; Harris & Hirsch, 2020; Holm, 2010; Krouse, 2007; Marston, 1972; Nalewicki, 2021; NICOA, 2019; Viola, 2008; Wooda et al., 2021).

This study …

Increasing Pediatric Asthma Follow-Up Visits And Decreasing Emergency Department Visits In A Primary Care Setting, Rima M. Basal

Doctor of Nursing Practice (DNP) Scholarly Projects - Archive

The implementation of the asthma follow-up visits bundle in the pediatric primary care setting is an evidence-based practice project aimed at improving pediatric asthma management. This manuscript discusses the deployment of the bundle, which includes pre- and post- knowledge assessment for parents/caregivers, asthma education through a short video and handout, administration of the asthma control test (ACT), completion of a written asthma action plan (WAAP), and education about the patient portal and telehealth. The goal is to improve parent/caregiver knowledge, treatment adherence, follow-up visit compliance (in-person or virtual) and reduce emergency department (ED) utilization. Parent/caregiver knowledge increased by 23%, ACT …

Advancing Machine Learning Approaches Through Robust Methodologies In Llm Code Generation, Adversarial Text Classification, And Unsupervised Learning, Anahita Samadi

Computer Science and Engineering Dissertations - Archive

This dissertation combines insights across text, code, and image modalities to advance the robustness, efficiency, and adaptability of machine learning models. Specifically, we address challenges like adversarial vulnerability in text, the impact of test strategies on code generation, and dimensionality in image representation in unsupervised learning domain. These efforts highlight pipelines for designing machine learning systems that are not only efficient, but also adaptable to complex environments. In addition, these efforts together help form the basis for a multimodal AI capable of thriving in medical applications that this dissertation prototypes for future efforts.

Parsimony, Parallelism, And Presentation: The Impact Of Presentation On Evaluation, Peter Mclemore

Psychology Theses - Archive

In professional and academic settings, support for creative ideas often hinges on how effectively those ideas are communicated in writing. This study examined how two sentence-level rhetorical features—parsimony and parallelism—influence the evaluation, support, and recall of written proposals featuring creative restaurant concepts. Using a 2×2 within-subjects design, 211 undergraduates reviewed proposals manipulated to vary in parsimony (high vs. low) and parallelism (high vs. low). Participants rated each proposal on quality, originality, elegance, support, and completed a recall task. Contrary to expectations, proposals with low parsimony and low parallelism received the highest evaluations across all three creative indicators (i.e., quality, originality, …

Enhancing Professional Development For Third Space Roles: Reflections On The Added Value Of Learning Circles, Linde Moriau, Réka Matolay, Emma Mckenna, Andrea Toarniczky, Judit Gáspár, Márta Frigyik, Catherine Bates

Articles

This paper explores how Learning Circles can support higher education staff in navigating the complexities of contemporary educational landscapes with greater confidence, creativity, and efficacy. It presents qualitative research rooted in the three-year Erasmus+ project CIRCLET, involving five European universities. Learning Circles were implemented as a core component of professional development programmes, aiming to foster quality integration of Community Engaged Research and Learning (CERL) in academic practice. Drawing on a collaborative autoethnographic research design, we reflected on our experiences as Learning Circle designers, facilitators, and participants to identify manifestations of professional growth emanating from our project activities. We build on …

Case Study Examining The Effects Of Mentorship On The Attrition Of Novice Principals In Urban School Districts, Nikia Barnett

Educational Leadership & Policy Studies Dissertations - Archive

The evolving role of school principals has undergone a significant transformation, shifting from traditional teacher management to a multifaceted leadership position that balances managerial and instructional responsibilities with a primary focus on enhancing student achievement. As principals now face increasingly complex challenges, including heightened school accountability, widening student achievement gaps, diminished community support, and fluctuating parent engagement, state programs and school districts struggle to provide adequate support to ensure the retention and longevity of these pivotal educational leaders in their roles.

This study examined how eight novice principals (with 0–5 years of experience) in urban school districts perceived mentorship influenced …

The Genesis Of Constitutions: A Natural Language Processing Approach, Tejas Ramdas, Patrick Chung-Chia Huang, Nuno Garoupa, Martin T. Wells, Yun-Chien Chang, Tom Ginsburg

Public Law and Legal Theory Working Papers

The diffusion of constitutional ideas and models has been the subject of extensive scholarship. In this paper, we use natural language processing methods to examine the persistence of constitutional ideas and models, focusing on how 16 core and iconic constitutions influence 572 newer constitutions over a long period (1900–2020). Our findings indicate that “old” core constitutions (such as those of the U.S. of 1789, Spain of 1812, France of 1848, and to a lesser extent France of 1791) have maintained significant influence over those 121 years. Austria of 1920 and France of 1946 are the most influential among the 20th …

Resetting The Imposter Phenomenon Reality: A Mixed Methods Research Study Of An Immediate Cognitive Behavioral Intervention On Stem Practitioners, Peter D. Smith

Dissertations

Imposter Phenomenon (IP) remains an essential framework for understanding how professionals attribute success to chance or misconceptions rather than genuine abilities. IP is especially pertinent for individuals who feel like frauds when attaining higher positions, convinced the promotions resulted from luck rather than accomplishments and credibility. The misguided mindset cultivates feelings of inadequacy or deceit, even when accolades, experiences, and qualifications prove the contrary. The prevalence of IP is widespread, impacting numerous individuals who have faced or continue to endure episodes of varying intensity. The psychological state makes individuals vulnerable to feelings of inadequacy, leading to reduced productivity, increased procrastination, …

No Longer Invisible: 2025 Massachusetts Asian And Pacific Islander American (Apia) Community Survey Report, Watanabe P. Dhaurali S.

Institute for Asian American Studies Publications

The first Massachusetts Asian Pacific Islander American (APIA) Community Survey gathered opinions from 1,414 APIAs across the state in March 2025 on issues such as economic stability, political engagement, safety and sense of belonging, healthcare access, as well as recent executive actions by the federal administration. The survey was offered in Chinese, Vietnamese, and Khmer languages in addition to English.The resulting data was disaggregated by the three largest APIA subgroups in MA: Chinese, Indian and Vietnamese Americans.
Key findings include that housing affordability and cost of living are the top economic concerns for APIAs across Massachusetts; both citizens and non-citizens …

The Internist's Post V.1 N.11, The University Of Texas Rio Grande Valley. School Of Medicine, Eunbee Cho, Osiris Fernandez, Hector D. Preciado, Sobia Memon, Diana Acosta Pozo, Corey Haines, Roque Mifuji Lira, Martha Solis, Cesar Gutierrez

School of Medicine Publications

No abstract provided.

Deep5mc: Predicting 5-Methylcytosine (5mc) Methylation Status Using A Deep Learning Transformer Approach, Evan Kinnear, Houssemeddine Derbel, Zhongming Zhao, Qian Liu

Faculty, Staff and Student Publications

DNA methylations, such as 5-methylcytosine (5mC), are crucial in biological processes, and aberrant methylations are strongly linked to various human diseases. Genomic 5mC is not randomly distributed but exhibits a strong association with genomic sequences. Thus, various computational methods were developed to predict 5mC status based on DNA sequences. These methods generated promising achievements and overcome the limitations of experimental approaches. However, few studies have comprehensively investigated the dependency of 5mC on genomic sequences, and most existing methods focus on specific genomic regions. In this work, we introduce Deep5mC, a deep learning transformer-based method designed to predict 5mC methylations. Deep5mC …

Generalizable Self-Supervised Learning For Brain Cta In Acute Stroke, Yingjun Dong, Samiksha Pachade, Kirk Roberts, Xiaoqian Jiang, Sunil A Sheth, Luca Giancardo

Faculty, Staff and Student Publications

Acute stroke management involves rapid and accurate interpretation of CTA imaging data. However, generalizable models for multiple acute stroke tasks able to learn from unlabeled data do not exist. We propose a linear probed self-supervised contrastive learning utilizing 3D CTA images and the findings section of radiologists' reports for pretraining. Subsequently, the pretrained model was applied to four disparate tasks: large vessel occlusion (LVO) detection, acute ischemic stroke detection, acute ischemic stroke, intracerebral hemorrhage classification, and ischemic core volume prediction. The tasks chosen are particularly challenging as they cannot be directly extracted from the radiology reports findings with keywords. The …

Enhancing Cross-Domain Generalizability In Social Determinants Of Health Extraction With Prompt-Tuning Large Language Models, Cheng Peng, Zehao Yu, Kaleb E Smith, Wei-Hsuan Lo-Ciganic, Jiang Bian, Yonghui Wu

Faculty, Staff and Student Publications

The progress in natural language processing (NLP) using large language models (LLMs) has greatly improved patient information extraction from clinical narratives. However, most methods based on the fine-tuning strategy have limited transfer learning ability for cross-domain applications. This study proposed a novel approach that employs a soft prompt-based learning architecture, which introduces trainable prompts to guide LLMs toward desired outputs. We examined two types of LLM architectures, including encoder-only GatorTron and decoder-only GatorTronGPT, and evaluated their performance for the extraction of social determinants of health (SDoH) using a cross-institution dataset from the 2022 n2c2 challenge and a cross-disease dataset from …

Automatic Summarization Of Doctor-Patient Encounter Dialogues Using Large Language Model Through Prompt Tuning, Mengxian Lyu, Cheng Peng, Xiaohan Li, Patrick Balian, Jiang Bian, Yonghui Wu

Faculty, Staff and Student Publications

Automatic text summarization (ATS) is an emerging technology to assist clinicians in providing continuous and coordinated care. This study presents an approach to summarize doctor-patient dialogues using generative large language models (LLMs). We developed prompt-tuning algorithms to instruct generative LLMs to summarize clinical text. We examined the prompt-tuning strategies, the size of soft prompts, and the few-short learning ability of GatorTronGPT, a generative clinical LLM developed using 277 billion clinical and general English words with up to 20 billion parameters. We compared GatorTronGPT with a previous solution based on fine-tuning of a widely used T5 model, using a clinical benchmark …

Enhancing Medical Coding Efficiency Through Domain-Specific Fine-Tuned Large Language Models, Zhen Hou, Hao Liu, Jiang Bian, Xing He, Yan Zhuang

Faculty, Staff and Student Publications

Medical coding is essential for healthcare operations yet remains predominantly manual, error-prone (up to 20%), and costly (up to $18.2 billion annually). Although large language models (LLMs) have shown promise in natural language processing, their application to medical coding has produced limited accuracy. In this study, we evaluated whether fine-tuning LLMs with specialized ICD-10 knowledge can automate code generation across clinical documentation. We adopted a two-phase approach: initial fine-tuning using 74,260 ICD-10 code-description pairs, followed by enhanced training to address linguistic and lexical variations. Evaluations using a proprietary model (GPT-4o mini) on a cloud platform and an open-source model (Llama) …

Not Fully Synthetic: Llm-Based Hybrid Approaches Towards Privacy-Preserving Clinical Note Sharing, Atiquer Rahman Sarkar, Yao-Shun Chuang, Xiaoqian Jiang, Noman Mohammed

Faculty, Staff and Student Publications

The publication and sharing of clinical notes are crucial for healthcare research and innovation. However, privacy regulations such as HIPAA and GDPR pose significant challenges. While de-identification techniques aim to remove protected health information, they often fall short of achieving complete privacy protection. Similarly, the current state of synthetic clinical note generation can lack nuance and content coverage. To address these limitations, we propose an approach that combines de-identification, filtration, and synthetic clinical note generation. Variations of this approach currently retain 36%-61% of the original note's content and fill the remaining gaps using an LLM, ensuring high information coverage. We …

Leveraging Gpt-4o For Automated Extraction Of Neural Projections From Scientific Literature, Rashmie Abeysinghe, Gorbachev Jowah, Licong Cui, Samden D Lhatoo, Guo-Qiang Zhang

Faculty, Staff and Student Publications

Sudden Unexpected Death in Epilepsy (SUDEP) is a major cause of death for epilepsy patients having uncontrolled seizures. Understanding the complex neural circuits within the central nervous system is crucial for understanding the mechanisms underlying cardiorespiratory regulation, particularly in the context of SUDEP. This study explores the potential of GPT-4o, a cutting-edge language model, to automate the extraction of neural projections from scientific literature. We developed prompts to extract neuroscientific structures, extract projections, and perform synonym harmonization. Applying the approach to four neuroscientific articles, the method extracted 205 projections. A random sample of 100 projections identified was handed over to …

Explainable Diagnosis Prediction Through Neuro-Symbolic Integration, Qiuhao Lu, Rui Li, Elham Sagheb, Andrew Wen, Jinlian Wang, Liwei Wang, Jungwei W Fan, Hongfang Liu

Faculty, Staff and Student Publications

Diagnosis prediction is a critical task in healthcare, where timely and accurate identification of medical conditions can significantly impact patient outcomes. Traditional machine learning and deep learning models have achieved notable success in this domain but often lack interpretability which is a crucial requirement in clinical settings. In this study, we explore the use of neuro-symbolic methods, specifically Logical Neural Networks (LNNs), to develop explainable models for diagnosis prediction. Essentially, we design and implement LNN-based models that integrate domain-specific knowledge through logical rules with learnable weights and thresholds. Our models, particularly Mmulti-pathway and Mcomprehensive, demonstrate superior performance over traditional models …

Application Of Artificial Intelligence Techniques To Improve Leadership Decision Making With Uncertainty, Michael David Parrish

Doctoral Dissertations

"Every good leader is a good manager, but not every good manager is a good leader. The difference between the leader and the manager is critical decision-making. Today’s decision-making environment is characterized as Volatile, Uncertain, Complex, and Ambiguous (VUCA). With the exponential increase in the technical capabilities of systems, the human has become the weakest link in the use of such systems. To remain relevant, good leaders must continuously adapt to new advances in technology and processes.

The research contributions of this work provide several unique and novel solutions for leaders to utilize artificial intelligence tools to improve and optimize …

Parallel Algorithms For Large Scale Dynamic Graph Analysis, Arindam Khanda

Doctoral Dissertations

A complex system of interacting entities in contemporary scenarios, be it biological, technological, or social, can be represented using graphs. Dynamic graphs, unlike their static counterparts, are ones in which the underlying topology changes over time. These networks act as a model for numerous systems, from transportation to social interactions, capturing the ever-evolving nature of real-world phenomena. However, the inherent temporality of these networks presents a unique set of challenges and the traditional static graph algorithms often fall short in efficiency and applicability. In our research, we delve into the complexities presented by large dynamic networks and suggest various methodologies …

Speech Pattern Disorders In Verbally Fluent Individuals With Autism Spectrum Disorder: A Machine Learning Analysis, Chuanbo Hu, Jacob Thrasher, Wenqi Li, Mindi Ruan, Xiangxu Yu, Lynn K Paul, Shuo Wang, Xin Li

2020-Current year OA Pubs

INTRODUCTION: Diagnosing Autism Spectrum Disorder (ASD) in verbally fluent individuals based on speech patterns in examiner-patient dialogues is challenging because speech-related symptoms are often subtle and heterogeneous. This study aimed to identify distinctive speech characteristics associated with ASD by analyzing recorded dialogues from the Autism Diagnostic Observation Schedule (ADOS-2).

METHODS: We analyzed examiner-participant dialogues from ADOS-2 Module 4 and extracted 40 speech-related features categorized into intonation, volume, rate, pauses, spectral characteristics, chroma, and duration. These acoustic and prosodic features were processed using advanced speech analysis tools and used to train machine learning models to classify ASD participants into two subgroups: …

A National Zoning Atlas To Inform Housing Research, Policy, And Public Participation, Sara C. Bronin, Wenfei Xu, Scott Markley, Diana Drogaris

GW Law Faculty Publications & Other Works

Through a unique combination of data science and legal analysis techniques, the National Zoning Atlas is creating the first public, online repository of standardized data about zoning. This article first discusses the context for and methodology behind the atlas. It then establishes three possibilities for using the atlas, including facilitating research (including fair housing research), strengthening planning, and empowering the public.

Overcoming Racial Harms To Democracy From Artificial Intelligence, Spencer A. Overton

GW Law Faculty Publications & Other Works

While the United States is becoming more racially diverse, generative artificial intelligence and related technologies threaten to undermine truly representative democracy. Left unchecked, AI will exacerbate already substantial existing challenges, such as racial polarization, cultural anxiety, antidemocratic attitudes, racial vote dilution, and voter suppression. Synthetic video and audio (“deepfakes”) receive the bulk of popular attention—but are just the tip of the iceberg. Microtargeting of racially tailored disinformation, racial bias in automated election administration, discriminatory voting restrictions, racially targeted cyberattacks, and AI-powered surveillance that chills racial justice claims are just a few examples of how AI is threatening democracy. Unfortunately, existing …