Digital Commons Network^™

Ml Model To Better Identify Instances Of Bullying Faced By Members Of The Lgbtq+ Community, Arslan Bisharat

Master's Theses

Cyberbullying poses a significant threat to online communities, with the LGBTQ+ community facing disproportionately higher rates of harassment. While existing cyberbullying detection systems have made progress in identifying general instances of online harassment, they often fail to capture the nuanced and context-dependent nature of LGBTQ+-targeted bullying. This thesis presents a novel approach to this challenge by developing SpectrumNet, an LGBTQ+-centric transformer-based model for cyberbullying detection. Our research was conducted in two phases. In Phase 1, we evaluated the effectiveness of pre-trained transformer models (RoBERTa, BERT, and GPT-2) in identifying LGBTQ+-related cyberbullying. Building on these findings, Phase 2 introduced SpectrumNet which …

Skinned Worlds: Virtual Reality And The Postbiological, Alicia Corts

Research, Publications & Creative Work

This paper explores the concept of “skinning” in educational virtual reality (VR) environments, reframing it as a dramaturgical and pedagogical act rather than a mere visual overlay. Through comparative analysis of The World of Hugo Simberg and C!ao: Piazza Navona Experience, the study investigates how design decisions shape user identity, agency, and the performance of learning. Drawing on performance theory, it examines how avatars – or their absence – guide learners through curated paths, subtly influencing perception and interaction. While both applications strip away overt self-representation, they open space for subversion, reflection, and imaginative engagement. The paper argues that skinning …

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 …

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 …

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) …

“Fish In Different Water, To Survive Or To Thrive?” Understanding Direct-Entry Pre-Sessional Chinese Students’ Learning Experiences During Covid-19 And Its Aftermath, Lingyuan Osgathorpe

School of Law, Humanities and Social Sciences Theses

This thesis explores the learning experiences of direct-entry pre-sessional Chinese students in the UK during the COVID-19 pandemic and its aftermath, using Pierre Bourdieu’s theoretical framework. Through the metaphor of students as "fish in different water", the research examines how these learners navigate and adapt to new educational environments that often starkly contrast with their prior experiences.The study employs a two-stage semi-structured interview process with student participants, first in China during their online pre-sessional courses and later in the UK during their degree programmes. All interviews were conducted in Mandarin Chinese to ensure linguistic and cultural accessibility. Additionally, expert interviews …

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 …

Recovery Of Verbal Working Memory Depends On Left Hemisphere White Matter Tracts, Randi C Martin, Junhua Ding, Ali I Alwani, Steve H Fung, Tatiana T Schnur

Faculty, Staff and Student Publications

Researchers propose that the recovery of language function following stroke depends on the recruitment of perilesional regions in the left hemisphere and/or homologous regions in the right hemisphere. Many investigations of recovery focus on changes in gray matter regions, whereas relatively few examine white matter tracts and none address the role of these tracts in the recovery of verbal working memory. The present study addressed these gaps, examining the role of left versus right hemisphere tracts in the longitudinal recovery of phonological and semantic working memory. For 24 individuals with left hemisphere stroke, we assessed working memory performance within 1 …

The Tripod-Llm Reporting Guideline For Studies Using Large Language Models, Jack Gallifant, Majid Afshar, Saleem Ameen, Yindalon Aphinyanaphongs, Shan Chen, Giovanni Cacciamani, Dina Demner-Fushman, Dmitriy Dligach, Roxana Daneshjou, Chrystinne Fernandes, Lasse Hyldig Hansen, Adam Landman, Lisa Lehmann, Liam G Mccoy, Timothy Miller, Amy Moreno, Nikolaj Munch, David Restrepo, Guergana Savova, Renato Umeton, Judy Wawira Gichoya, Gary S Collins, Karel G M Moons, Leo A Celi, Danielle S Bitterman

Faculty, Staff and Student Publications

Large language models (LLMs) are rapidly being adopted in healthcare, necessitating standardized reporting guidelines. We present transparent reporting of a multivariable model for individual prognosis or diagnosis (TRIPOD)-LLM, an extension of the TRIPOD + artificial intelligence statement, addressing the unique challenges of LLMs in biomedical applications. TRIPOD-LLM provides a comprehensive checklist of 19 main items and 50 subitems, covering key aspects from title to discussion. The guidelines introduce a modular format accommodating various LLM research designs and tasks, with 14 main items and 32 subitems applicable across all categories. Developed through an expedited Delphi process and expert consensus, TRIPOD-LLM emphasizes …

Psychometric Properties And Meaningful Change Thresholds For The Qol-E Instrument In Patients With Myelodysplastic Neoplasms, Esther Natalie Oliva, Shien Guo, Jennifer Lord-Bessen, Aylin Yucel, Roberto Latagliata, Massimo Breccia, Giuseppe A Palumbo, Grazia Sanpaolo, Marta Riva, Valeria Santini, Uwe Platzbecker, Guillermo Garcia-Manero, Pierre Fenaux, Christopher G Pelligra

Faculty, Staff and Student Publications

peripheral blood cytopenias, and an increased risk of progression to acute myeloid leukemia. One of the main treatment goals is improving quality of life (QoL), particularly for patients with lower-risk MDS (LR-MDS) who may live longer with compromised QoL. The QOL-E© is a patient-reported outcome (PRO) measure specifically developed to address the lack of a health-related QoL questionnaire for patients with MDS. The objective of this study was to evaluate the psychometric performance of the QOL-E in patients with LR-MDS.

Methods: Data from four clinical trials in MDS (MEDALIST, DARB-MDS, EQoL-MDS, and RevMDS trials) were used to assess construct validity, …

Building Boundary-Crossers In Clinical-Translational Research: An Exploratory Study Of A Novel Communication Intervention, Carrie Cameron, Irene Rull-Garcia, Laura G Meyer, Jose-Armando Fernandez-Guerrero, Erin K Dahlstrom, Christine Bell, Christine Pribbenow

Faculty, Staff and Student Publications

INTRODUCTION: Integrating scientific research across multiple disciplines to advance breakthroughs is at the heart of clinical-translational science (CTS); among competencies that have been identified as essential for progress, skillful communication is critical. Few tools are available to address the social dynamics of the multidimensional diversity characteristics of CTS. We created the "Building a Diverse Biomedical Workforce Through Communication Across Difference (CAD)" workshop intervention. Based on principles of intercultural communication, CAD taught novel situationally-based communication skills to dyads of near-peer mentors and their undergraduate mentees. This study reports on the effectiveness of the operative mechanisms employed in CAD workshops for helping …

From Guesswork To Growth: Understanding How Preservice Teachers Interpret And Enact Appraisal Within A Teacher Performance Assessment, Allison Byth, Thembi Mason, Laurie Werrett

Australian Journal of Teacher Education

In Australia, the Graduate Teacher Performance Assessment (GTPA) is a capstone summative assessment evaluating preservice teachers ‘classroom readiness’ across five practices: planning, teaching, assessing, reflecting and appraising. In our work with preservice teachers, appraising the impact of teaching emerged as the most conceptually and practically challenging practice. To understand how preservice teachers interpreted and enacted appraisal, this exploratory case study analysed 15 GTPA submissions that met standards in the first four practices but not in appraisal. The findings highlight common issues, including conflating appraisal with general reflection, difficulty interpreting and applying data meaningfully, and presenting narrative accounts rather than evaluative …

Fostering Learner Autonomy In Efl Setting Through Multiliteracy Pedagogy: A Case In Tertiary Education, Desi Surlitasari Dewi, Rudi Hartono, Sri Wahyuni, Sri Lestari

Australian Journal of Teacher Education

The intersection of learner autonomy and multiliteracy pedagogy in EFL settings remains underexplored. To address this gap, this study investigates how multiliteracy pedagogy fosters learner autonomy in the EFL context. The study involved 24 university students majoring in English, for whom English was primarily a third language. The participants attended an intermediate speaking course, a compulsory class in their curriculum. A learner autonomy questionnaire was administered to measure autonomy in the EFL speaking context before and after the implementation. Additionally, semi-structured interviews and classroom observations enriched the data. The findings indicate a significant improvement in learner autonomy following the implementation …

Raising Information Literacy Of Biology Pre-Service Teachers: Study On Three Problem Solving Methods, Muhammad Saefi, Hadi Suwono, Rifka Fachrunnisa, Widi Cahya Adi, Herawati Susilo, Ahmad Kamal Sudrajat

Australian Journal of Teacher Education

Integrating information literacy teaching methods with subject-content curricula to develop prospective biology teacher students' abilities is still problematic. This quasi-experimental research, with a nonrandomized pretest-posttest control group design, intends to compare the influences of challenge-based learning (CBL), problem-based learning (PBL), and problem-solving through classroom discussion (PCD) – as the control- in improving information literacy. This study involved 183 students from three classes of freshmen and three sophomores in the biology education program. The results showed that CBL significantly improved students' information literacy compared to PBL and PCD across five indicators: 1) identifying the required information, 2) attaining the required information …

Teacher Resilience In South Africa: The Self-Efficacy And Teacher Efficacy Of Pre-Service Teachers Preparing To Teach In A Challenged Context, Carine Jonker, Marien Alet Graham, Liesel Ebersöhn

Australian Journal of Teacher Education

This publication addresses a gap in high-quality, quantitative studies on teacher resilience within the less-researched context of the Global South. Specifically, it examines the self-efficacy and teacher efficacy of three cohorts of final-year pre-service teachers in South Africa (N = 1,193). This article aimed to contribute evidence on teacher resilience from an under-researched African context and population. Despite contextual challenges, the pre-service teachers indicated high intrapersonal resilience-enabling pathways, with a statistically significant relationship between their self-efficacy and teacher efficacy. The authors propose an evidence-based framework of how high levels of self-efficacy and teacher efficacy may enable prospective teachers to teach …

A Qualitative Exploration Of Changes In English Language Teachers’ Perceptions Of Their Agency Through Engaging In Action Research, Khadije Aqajani Delavar, Mohammad R. Hashemi, Mohammad Nabi Karimi

Australian Journal of Teacher Education

While research on language teachers’ engagement in action research (AR) has received a surge of attention, the influence of AR programs on English language teacher agency remains under-examined. This qualitative study explored how ten English language teachers’ perceptions of their agency changed as they participated in an action research engagement program (AREP). Data were collected using semi-structured interviews and reflective journals. The findings revealed that participation in the AR program changed the teachers’ perceptions of their agency and empowered them to enact higher levels of agency in their classroom practice. The study also highlights the important role of institutional support …

A Hybridized Cnn-Lstm-Mlp-Knn Model For Short-Term Solar Irradiance Forecasting, Doaa El-Shahat, Ahmed Tolba

Sustainable Machine Intelligence Journal

The accurate prediction of solar irradiance is essential for maximizing renewable energy production. Despite some researchers struggling to achieve sufficient prediction accuracy and model quality, we are working on developing a forecasting model for solar radiation using advanced artificial intelligence techniques. Solar energy offers a sustainable alternative to fossil fuels and has a wide range of applications, making the development of an effective forecasting model crucial. This paper proposes a hybrid model that combines deep learning and machine learning methods, denoted as CNN-LSTM-MLP-KNN. By combining Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Multilayer Perceptron (MLP) and K-Nearest Neighbors (KNN), …

Homelands Not Graveyards, M. Alexander Pearl

Faculty Articles

Within the last five years, the U.S. Supreme Court has taken up several transformative cases affecting Native nations and federal Indian law jurisprudence. The Supreme Court’s decision in Arizona v. Navajo Nation is no different. This Article examines that decision and situates it within that legal history as well as the realities of present-day water resource availability. While recent decisions have shown the Court’s willingness to confirm fundamental components of federal Indian law, such as the legal tests for determining the persistence of reservation and the authority of Congress to enact the Indian Child Welfare Act, the Court’s examination of …

Learning From Mistakes: A Quantitative Comparative Study Of Court Decisions Involving The Excuse Of Contractual Mistake, Brian Mccall

Faculty Articles

In almost two decades of teaching Contracts law to first year law students, I have found one of the hardest doctrines to teach to be the excuse of mistake, both mutual and unilateral Students typically express the frustration that they cannot articulate a clear standard to determine when a party’s claim of an excusing mistake will succeed or fail. This frustration with the doctrine seems to be thousands of years old. After years of struggling to teach this topic, I resolved to dig deeper into the caselaw to see if any trends could be observed that would help dispel some …

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 …

Supporting Students In Transition To And From High School, David Naff, Jamelle Simmons, Meleah Ellison, Shelby Garofalo, Laura Thoene, Kim Dupre, Sooyeon Moon, Karli Johanson, Paul Harris, Martin Darby, William L. Parker, Tracy Booker, Janel Miley, Mysia Perry, Xun Liu, Maggie Szydlowski, Ahmed Abbas

MERC Publications

This report is part of a two-part series focused on supporting students in transition across PK-12 education. Specifically, it explores transitions into and out of high school school. The report includes a synthesis of recent peer-reviewed literature, analysis of publicly available data (with a specific focus on Virginia and the metropolitan Richmond region), and summary of relevant federal, state, and school division-level policies. It answers the following questions:

1. What are key considerations for supporting students in transition to and from high school in the MERC region?

2. How do federal, state, and division policies guide support for student transitions …

Diet And Migration In The Prehistoric Populations In The Greater Nicoya Region Of Nicaragua, Ashley Nichols Whitten

Theses and Dissertations--Anthropology

The goal of this three-paper dissertation was to engage with the archaeological hypotheses of prehistoric migration from Mesoamerica into the Greater Nicoya Region of Nicaragua. To that aim, this dissertation contains a discussion on the current information supporting and refuting the migratory hypotheses. New data is also presented in the form of stable isotope analysis and bioarcheological analysis of dentition from prehistoric archaeological sites in Pacific Nicaragua.

The first paper titled “Prehistoric Foodways and Migration Hypotheses in the Greater Nicoya Region of Nicaragua” outlines the current available research concerning migration out of Mesoamerica into Pacific Nicaragua. The potential migration is …

Computational Bridges: Enhancing Natural Language Processing Of Swahili., Joyce Murungi

Harrisburg University Other Works

Swahili remains significantly underrepresented in natural language processing (NLP) despite being one of the most widely spoken languages in Africa. Computational Bridges: Enhancing Natural Language Processing of Swahili addresses this gap through computational linguistics, corpus creation, and large-scale analysis of Swahili syntax and lexical structure. Central to this study is GUMZO, a novel corpus developed from spontaneous conversational data collected from YouTube videos, television panel discussions, political speeches, religious discourse, and unscripted broadcasts. Unlike many existing datasets that rely on formal or translated text, GUMZO captures authentic language use and provides a stronger foundation for NLP research involving low-resource languages. …

A Power Dependence Model Of The Impact Of Leader Impostorism On Supervisor Support And Undermining: The Moderating Role Of Power Distance, Xueqi Wen, Zihan Liu, Feng Qiu, Keith Leavitt, Xingyu Wang, Ziyang Tang

Research Collection Lee Kong Chian School Of Business

Leaders, often perceived as possessing exceptional confidence and competence, are not immune to feelings of self-doubt. Leader impostorism describes the experience that one’s attributes, experiences, skills, and abilities fall short of the standards expected in the leadership role, resulting in a sense of deception in fulfilling leadership responsibilities. While existing research has examined the antecedents and individual outcomes of leader impostorism, its implications for leaders’ treatment of subordinates remain largely unexplored. In this research, we investigate the downstream consequences of leader impostorism on behaviors directed toward subordinates. Integrating research on leader impostorism with power dependence theory, we propose that for …

Recognition Of Alzheimer’S Disease Stages Via Inceptionv3 And Resnet50, Iman Aljubouri, Mostafa Ragheb, Mohamad Hamady

Mesopotamian Journal of Big Data

Early and precise detection of Alzheimer’s disease (AD) is essential for successful treatment. This research presents a system that autonomously detects and categorizes the phases of Alzheimer’s disease via brain scans and sophisticated deep learning techniques, including InceptionV3 and ResNet50. These models started with pretrained weights and were augmented by including bespoke classification layers, which consisted of dropout, batch normalization, and dense layers to increase performance and mitigate overfitting. The preprocessing processes included scaling the picture to 224 by 224 pixels, using average filtering for denoising, and converting the color space to guarantee compatibility with the models. Evaluations of the …

Popular Originalism, Thomas G. Donnelly

Law Faculty Publications

Scholars are obsessed with the Roberts Court’s recent turn to history and tradition. But another source of authority has also emerged as an important feature of the Roberts Court’s approach to constitutional cases. In previous work, I have referred to this source of authority as popular meaning. While original meaning identifies the best reading of the Constitution’s text at the time of its ratification, popular meaning draws on sources of authority outside the courts to capture the constitutional views of the American people today. In this Article, I demonstrate how interpreters might draw on the Constitution’s popular meaning to help …

Ai In Education: Teaching, Learning, And Thinking With Machines, Mary Strawderman, Jesse Senechal, Eric Ekholm, Jason Vest, Moe Greene, Dawn Johnson, Daniel Case, Oscar Keyes, Yaoying Xu, Robert Godwin-Jones, Kait Donahue, Kim Elmore, Haley Hendershot, Amy Corning, Paulina Guerrero, Samaher Aljudaibi, Justin Thurston

AI in Education: Teaching, Learning, and Thinking with Machines

This collection from the Humanities Research Center (HRC) Digital Technology in Education (DTE) working group arose from questions around how rapidly emerging generative AI tools are reshaping teaching and learning in K-12 and higher education. The contributions, loosely organized around ethics, policy, and practice, adopt a humanities lens to probe how generative AI and other emerging digital technologies impact human values, experiences, and social structures, surfacing their philosophical, ethical, cultural, and societal implications for schooling. From the perspectives of practitioners and interdisciplinary scholars, the pieces showcase various dimensions of AI's creative potential in education, while also questioning its limitations and …

A Bounded Custom Gpt Structures Operational Knowledge For Small-Industry Decision Support, Ikhwan Arief, Alizar Hasan, Nilda Tri Putri, Hafiz Rahmann

Knowledge Engineering and Data Science

Small manufacturing and craft-based firms increasingly use Generative Artificial Intelligence (GenAI) through public chat interfaces, low-cost tools, and informal experimentation. However, these firms often make operational decisions with incomplete records, tacit owner knowledge, fragmented spreadsheets, and limited managerial capacity. Under such conditions, open-ended chatbots may generate fluent but unsafe recommendations by overlooking missing information, contradictory evidence, feasibility constraints, and implementation constraints. This study presents Asisten Cerdas Industri Kecil as a bounded Custom GPT artifact for operational diagnosis, priority selection, and short-horizon action planning in small industries. Using a Design Science Research approach, the study develops a documented artifact corpus comprising …

Pedagogical Idealism, Race, And The Specter Of Africa In Josefina Aldecoa’S Historia De Una Maestra And Donato Ndongo’S Las Tinieblas De Tu Memoria Negra, Parker Lawson

Funded Scholarship

This article examines two contemporary Spanish novels, Josefina Aldecoa’s Historia de una maestra (1990) and Donato Ndongo’s Las tinieblas de tu memoria negra (1987), to assess the status of pedagogical idealism and experimentation with particular attention to Equatorial Guinea. Where previous scholarly attention to Aldecoa’s novel focuses on the protagonist’s credentials as a progressive pedagogue, this article draws attention to Gabriela’s etapa guineana to argue that the text cripples under the weight of postcolonial analysis as regards race, religion, class, and language. Drawing the novel into comparison with Ndongo’s text, the article argues that Spanish colonial pedagogy flattened, or sought …

A Morphosyntactic Study Of Arab American Code-Switching, Qamar Bazar

Publications and Research

This study examines the morphosyntactic behavior of Arabic lexical items in English-dominant speech among Arab American bilinguals. Based on original survey data from 22 participants between the ages of 16 and 32, the paper focuses on specific Arabic words, such as yani, khalas, and ana, in English conversations across informal and professional settings. The analysis shows that these lexical items function as free morphemes and are inserted without disrupting English syntax, often serving discourse, emotional, or identity-marking purposes. Grounded in sociolinguistic theory and Poplack’s model of intra-sentential code-switching, this paper argues that code-switching among Arab Americans is intentional and …