Digital Commons Network^™

Multitask Learning For Named Entity Recognition And Relationship Extraction, Adrienne D. Hembrick

Theses and Dissertations

Information Extraction (IE) is a fundamental task in Natural Language Processing (NLP), involving the identification of structured information from unstructured text. Two core components of IE—Named Entity Recognition (NER) and Relation Extraction (RE)—are widely used to extract key concepts and the relationships between them across various domains. However, the sequential dependency of RE on the output of NER makes it vulnerable to error propagation: inaccuracies in entity recognition can negatively affect downstream relation extraction.

To mitigate this issue, Multitask Learning (MTL) has been proposed as an approach that jointly models NER and RE, aiming to improve overall performance and reduce …

Leveraging Counterfactuals For Enhanced Natural Language Inference: A Multitask Knowledge Distillation Approach, Brian J. Davis

Browse all Theses and Dissertations

Natural-language inference (NLI) asks whether a hypothesis is entailed by, contradicts, or is neutral with respect to a premise. Modern transformers reach high raw accuracy on benchmarks such as SNLI, MNLI, and ANLI, yet they often rely on brittle lexical shortcuts and provide little insight into their decision process. This thesis shows that counterfactual-augmented knowledge distillation can simultaneously boost robustness and supply faithful, token-level explanations—without scaling model size. Four T5-v1_1 students (60M, 220M, 770M, 3B parameters) are trained under four curricula: (1) standard fine-tuning, (2) fine-tuning with free-text rationales, (3) multi-task distillation with naive counterfactuals, and (4) multi-task distillation with …

Reducing Operator Training Time Through Virtual Reality: A Case Study On The Lpkf Protomat E44 Machine, Joshua C. Patel

Browse all Theses and Dissertations

This thesis presents the development of an immersive virtual reality (VR) simulation that replicates the operation of the LPKF ProtoMat E44 PCB milling machine. Aimed at reducing operator training time and improving procedural understanding, the simulation offers an interactive and realistic environment where users can safely engage with machine workflows and start-up sequences. The emphasis is on accurate representation, usability, and maintaining immersion to support intuitive learning. Although formal evaluation is outside the scope of this work, the system is designed to serve as a foundation for cost-effective, scalable training in technical and manufacturing contexts, offering a modern alternative to …

Patient Subset Classification Using Encoded Embeddings And Knowledge Graph Retrieval-Augmented Generation, Benjamin A. Holmes

Browse all Theses and Dissertations

The widespread adoption of electronic medical records has created a vast reservoir of clinical data that can be leveraged to better understand how interventions relate to patient outcomes. Much of this information, however, exists as unstructured free-text, posing significant challenges for traditional statistical and machine-learning methods. Solving these challenges would allow the extraction of specific patient subpopulations (clinically relevant cohorts of individuals who share overlapping symptoms, risk factors, or diagnostic criteria), which could be used in precision medicine. Despite this promise, extracting these subpopulations from unstructured medical notes is an ongoing challenge due to the variability of clinical language and …

Kadysnet: A Novel Approach To Dyslexia Prediction In Children: Combining Handwritten Text Recognition With A Hybrid Cnn-Bilstm-Ctc Model And Personalized Learning Strategies, Shailesh Prabhakar Patil, Ravindra Sadashivrao Apare, Ravindra Honaji Borhade, Parikshit Narendra Mahalle

ASEAN Journal on Science and Technology for Development

This study introduces a novel approach for early dyslexia detection in children through automated handwriting analysis, integrating a hybrid CNN-BiLSTM-CTC architecture with personalized learning strategies. Our method combines a custom CNN-BiLSTM-CTC model with tailored educational interventions to support dyslexic learners. We analyzed children's handwritten text images in English, collected from specially designed tests involving word rewriting, sentence reconstruction, and paragraph composition, particularly challenging tasks for dyslexic individuals. Notably, our CNN-BiLSTM-CTC model achieved the best result with an accuracy of 97.67%, outperforming other architectures. Compared to pre-trained models like EfficientNetB7, DenseNet121, and MobileNetV2, our custom CNN-BiLSTM-CTC model demonstrated superior performance. Key …

Snake Optimization Algorithm For Fractional Order Controller In Three-Area Load Frequency Regulation, Shubra Goel, Omveer Singh

ASEAN Journal on Science and Technology for Development

The transition toward renewable energy-dominated power systems have accentuated the intricacies of frequency control, necessitating advanced regulatory mechanisms. This investigation articulates a tri-zonal frequency stabilization approach, employing a hybridized controller that synergizes Fuzzy Fractional-Order PI and Tilt-Integral-Derivative methodologies. The uniqueness of this approach is further amplified by the deployment of the Snake Optimization algorithm for precise parameter tuning. Set within a conventional grid topology integrated with assorted renewable energy sources, the study evaluates the controller’s adaptability and resilience through a series of comprehensive scenario-based analyses.

Enhancing Cybersecurity Through Autonomous Knowledge Graph Construction By Integrating Heterogeneous Data Sources, Hatoon Alharbi, Ali Hur, Hasan Alkahtani, Hafiz Farooq Ahmad

Research outputs 2022 to 2026

Cybersecurity plays a critical role in today’s modern human society, and leveraging knowledge graphs can enhance cybersecurity and privacy in the cyberspace. By harnessing the heterogeneous and vast amount of information on potential attacks, organizations can improve their ability to proactively detect and mitigate any threat or damage to their online valuable resources. Integrating critical cyberattack information into a knowledge graph offers a significant boost to cybersecurity, safeguarding cyberspace from malicious activities. This information can be obtained from structured and unstructured data, with a particular focus on extracting valuable insights from unstructured text through natural language processing (NLP). By storing …

Crop2cloud Platform: Real-Time Data Integration For Agricultural Water Monitoring, Bryan Nsoh, Abia Katimbo, Kendall Dejonge, Wei-Zhen Liang, Hongzhi N. Guo, Yufeng Ge, Derek M. Heeren, Yeyin Shi, Xin Qiao, Daran R. Rudnick, Hope Njuki Nakabuye, Birru Girma, Isa Kabenge, Joshua Wanyama

Department of Agricultural and Biological Systems Engineering: Faculty Publications

Efficient water management is vital for sustainable agriculture, yet integrating real-time data for precise irrigation remains a challenge. This study designed the Crop2Cloud (C2C) platform, a system that leverages advanced sensors using Internet of Things (IoT), edge and cloud computing techniques, and computed Water Stress Indices (WSIs) and machine learning models (i.e., fuzzy logic), to provide scalable and real-time irrigation decisions. The C2C platform aggregates several data including Volumetric Water Content (VWC) from TDR sensors (Acclima Inc., US) installed at four multiple depths, canopy temperatures (Tc) measured by Infrared Radiometers (IRTs) (Apogee Instruments, US), as well as weather information and …

Multi-Lingual And Cross-Domain Frontiers In Machine-Generated Content Detection, Gurunameh Singh Chhatwal

Theses and Dissertations (Comprehensive)

The rapid advancement of generative artificial intelligence, particularly Large Language Models (LLMs) such as GPT-4 and their multilingual capabilities, has significantly blurred the distinction between human-authored and machine-generated content. This technological evolution introduces critical challenges concerning the detection and attribution of textual authenticity and authorship, exacerbating societal issues like misinformation proliferation and compromising academic and professional integrity. Traditional detection methodologies, predominantly monolingual and heuristic-based, have demonstrated inadequate generalizability and efficacy against the sophisticated, multilingual capabilities of contemporary generative models.

This thesis addresses two major problems arising from these advancements. Firstly, it introduces novel multilingual detection methodologies explicitly designed to differentiate …

Investing In The Development Of The Next Generation Of Mch Leaders, Karen A. Mcdonnell, Jamal Percy, Lisa Anders, Monique J. Brown Ph.D., Mph, Alice R. Richman, Julianna Deardorff, Monica S. Ruiz, Jihong Liu Sc.D., Kelli Russell, Audrey Snyder, Cassondra Marshall

Faculty Publications

The public health landscape is constantly evolving to address the strengths and needs of the community. Training for the public health workforce is leading the way, establishing an ecosystem approach that integrates individuals within social, political, and environmental contexts to promote health equity within a framework of social justice. One area of public health that is innovatively preparing the next generation of leaders is maternal and child health (MCH). In the United States, key indicators of health disparities within MCH remain stagnant, highlighting the need for training programs that develop future MCH professionals from diverse backgrounds. These professionals will deliver …

The Integration Of Artificial Intelligence And Ontologies: Transformations In Knowledge Representation And Application, Grazia Serratore, Julaine Clunis

STEMPS Faculty Publications

Artificial Intelligence (AI) is reshaping the landscape of knowledge representation. There is an increasingly strong bidirectional relationship, between AI techniques and ontologies. AI techniques revolutionized traditional, manual ontology development and contribute to automated ontology construction, while ontologies enhance the performance of AI systems and their semantic accuracy. Through a comprehensive review of current literature, this paper aims to examine: i) how Machine Learning (ML) techniques contribute to the automated construction, refinement, and validation of ontologies; ii) the most widely used and effective ML approaches for ontology construction; iii) how domain-specific requirements influence the selection and adaptation of AI techniques for …

Overseas Agent Selection For International Logistics Companies: A Case Study, Hung-Ta Lin, Hua-An Lu

Journal of Marine Science and Technology–Taiwan

International logistics companies (ILCs) have aggressively participated in the supply chain systems of production industries as part of global trade development. They have adopted diversified and multinational models by expanding their overseas services to align with the various characteristics of regional economic patterns. Assigning specific local agents for their forwarding operations has been one of the key strategies for overseas services. Notably, the performance of these selected agents has significantly influenced service quality in their respective regions. In collaboration with the top management of an ILC, this study proposes an evaluation framework comprising 4 criteria and 16 subcriteria for selecting …

Exploring The Red Sea Crisis’S Supply Chain Disruption Impacts On The Shipping Industry, Po-Hsing Tseng, Nick Pilcher

Journal of Marine Science and Technology–Taiwan

Since October 7, 2023, the Houthi militia in Yemen has continuously attacked Israel and its nearby ships in the Red Sea. This crisis has impacted on global ship operations, freight rates, and shipping networks, since this conflict seriously affects the international trade between Europe and Asia and brings many unpredictable butterfly effects (e.g. supply chain disruption, and high transportation costs in shipping and air cargo). This papero explores its impact on the supply chain and suggests potential solutions;, and to consider perceptions on who should be responsible for the crisis, and who will actually shoulder the responsibility for the crisis. …

Cognitive Map Generation For Vision And Language Navigation, Alexander Sandoval Mesa

Dissertations and Theses

Visual-Language Navigation (VLN) presents significant challenges for autonomous agents, such as robots and virtual assistants, particularly in complex, dynamic environments where the seamless integration of visual perception and natural language understanding is critical. Traditional VLN systems often struggle with effectively aligning language instructions and visual scene understanding, limiting their adaptability and navigation efficiency.

This thesis proposes a novel Cognitive Map-based framework that addresses these challenges by transforming natural language navigation instructions into structured graph representations. The Cognitive Map consists of nodes representing waypoints, landmarks, decision points, and edges encoding spatial relationships and navigational actions. These maps are generated using Large …

Interpreting Topic Models In Byte-Pair Encoding Space, Jia Peng Lim, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

Byte-pair encoding (BPE) is pivotal for processing text into chunksize tokens, particularly in Large Language Model (LLM). From a topic modeling perspective, as these chunksize tokens might be mere parts of valid words, evaluating and interpreting these tokens for coherence is challenging. Most, if not all, of coherence evaluation measures are incompatible as they benchmark using valid words. We propose to interpret the recovery of valid words from these tokens as a ranking problem and present a model-agnostic and training-free recovery approach from the topic-token distribution onto a selected vocabulary space, following which we could apply existing evaluation measures. Results …

Adapting Installation Instructions In Rapidly Evolving Software Ecosystems, Haoyu Gao, Christoph Treude, Mansooreh Zahedi

Research Collection School Of Computing and Information Systems

files play an important role in providing installation-related instructions to software users and are widely used in open source software systems on platforms such as GitHub. Software projects evolve rapidly alongside their dependencies in dynamic software ecosystems, requiring frequent updates to installation instructions. These instructions are crucial for users to start with a software project. Despite their significance, there is a lack of systematic understanding regarding the documentation efforts invested in README files and the triggers behind them. To fill the research gap, we conducted a qualitative study, investigating 400 GitHub repositories with 1,163 README commits that focused on updates …

The Gradient Puppeteer: Adversarial Domination In Gradient Leakage Attacks Through Model Poisoning, Kunlan Xiang, Haomiao Yang, Meng Hao, Shaofeng Li, Haoxin Wang, Zikang Ding, Wenbo Jiang, Tianwei Zhang

Research Collection School Of Computing and Information Systems

In Federated Learning (FL), clients share gradients with a central server while keeping their data local. However, malicious servers could deliberately manipulate the models to reconstruct clients' data from shared gradients, posing significant privacy risks. Although such Active Gradient Leakage Attacks (AGLAs) have been widely studied, they suffer from two severe limitations: 1) coverage: no existing AGLAs can reconstruct all samples in a batch from the shared gradients; 2) stealthiness: no existing AGLAs can evade principled checks of clients. In this paper, we address these limitations with two core contributions. First, we introduce a new theoretical analysis approach, which uniformly …

A Study Of Pattern Of Cybercrime Abuse Of Individual Internet Users In Umuahia North Lga, Abia State Of South-Eastern Nigeria, Ogochukwu Favour Nzeakor, Rita Ngozi Okafor, Chibuike Ndubuisi Nwoke

International Journal of Cybersecurity Intelligence & Cybercrime

Although a number of studies exist on cybercrime and its abuses, little is known about the pattern of cybercrime abuses individual Internet users experience in Nigeria, especially the south eastern region. Using data collected via various methods, this study examines the pattern of cybercrime abuses of individual Internet users in Umuahia, Abia State, of South Eastern Nigeria. The result of the analysis of 1,067 samples drawn from 223,134 Internet users in Umuahia North LGA of Abia Sate showed that: while most users are victims of stolen ICT-gadgets (19%), fraud related offences (17%), and hacking (15%); they rarely fall victims of …

Fine-Tuned Large Language Models (Llms): Improved Prompt Injection Attacks Detection, Md Abdur Rahman, Hossain Shahriar, Guillermo Francia, Fan Wu, Alfredo Cuzzocrea, Sheikh Iqbal Ahamed

Computer Science Faculty Research and Publications

Large language models (LLMs) are becoming a popular tool as they have significantly advanced in their capability to tackle a wide range of language-based tasks. However, LLMs applications are highly vulnerable to prompt injection attacks, which poses a critical problem. These attacks target LLMs applications through using carefully designed input prompts to divert the model from adhering to original instruction, thereby it could execute unintended actions. These manipulations pose serious security threats which potentially results in data leaks, biased outputs, or harmful responses. This project explores the security vulnerabilities in relation to prompt injection attacks. To detect whether a prompt …

Contextual Embedding-Based Clustering To Identify Topics For Healthcare Service Improvement, K M Sajjadul Islam, Ravi Teja Karri, Srujan Vegesna, Jiawei Wu, Praveen Madiraju

Computer Science Faculty Research and Publications

Understanding patient feedback is crucial for improving healthcare services, yet analyzing unlabeled short-text feedback presents challenges due to limited data and domainspecific nuances. Traditional supervised approaches require extensive labeled datasets, making unsupervised methods more practical for extracting insights. This study applies unsupervised techniques to analyze 439 survey responses from a healthcare system in Wisconsin, USA. A keyword-based filter was used to isolate complaint-related feedback using a domain-specific lexicon. To identify dominant themes, we evaluated traditional topic models such as Latent Dirichlet Allocation (LDA) and Gibbs Sampling Dirichlet Multinomial Mixture (GSDMM) - alongside BERTopic, a neural embedding-based clustering method. To improve …

40 Years Passing By In (A) “Heartbeat”, Christopher Janney

Monsoon: South Asian Studies Association Journal

This article outlines the forty-year history go Christopher Janney’s iconic performance work “HeartBeat.” Trained as an architect and jazz musician, Janney combines these two disciplines to create large interactive light/sound installations and performances. On the one hand, he attempts to make architecture more like music, more “alive” using real-time sensing devices as on his “Harmonic Convergence” installation on the Miami International Airport. On the other hand, he attempts to make music more like architecture, more physical more visual. In his “HeartBeat” performance, a dancer wears a heartbeat device that amplifies his/her heartbeat in “real time.” Over this pulse, live musicians …

Yoga Sūtra Book Four And The Yogabindu, Christopher Key Chapple

Monsoon: South Asian Studies Association Journal

Four themes from the fourth pāda of Patañjali’s Yoga Sūtra are examined here: karma (IV:2-13), individuality (IV:16), humility (IV:20-22), and freedom (IV:29-34). Affirmations and variant interpretations of these four are provided from Haribhadra Virahāṅka’s Yogabindu, most likely composed a century or two later.

Collective Karma: From The Yogasūtra To Taylor Swift, Zoë Slatoff

Monsoon: South Asian Studies Association Journal

The Indian knowledge traditions speak at length about karma, which is divided into three categories—accumulated (saṃcita), ripe (prārabdha), and future (kriyamāṇa). While much is said about our individual karma and how it affects our repeated reincarnation in cyclic existence, how does this affect others and how do others affect us? In other words, as this paper explores, what is our reciprocal and collective karma?

Painting Liberation: A Visual Ontology Of The Interplay Of Mind And Consciousness In Book Four Of The Yoga Sūtras, Melissa Townsend

Monsoon: South Asian Studies Association Journal

This presentation explores how artistic practice can illuminate philosophical questions about the nature of mind, consciousness, and liberation as articulated in Book Four of the Yoga Sūtras of Patañjali. Through a series of paintings that visually translate these sūtras, I propose that art offers a unique, experiential means of engaging with the complex relationship between citta (mind) and puruṣa (consciousness).

The paintings function as a form of “visual ontology,” inviting viewers to engage with abstract philosophical concepts in a sensory, tangible way. By depicting the interplay between the mutable processes of the mind and the unchanging awareness of consciousness, these …

Synthesizing Deception: Countering Large Language Model-Generated Phishing Campaigns Through Adaptive Semantic Anomaly Detection, Bekim Fetaji, Debabrata Samanta

Mesopotamian Journal of Computer Science

The paper fills in a gap in the literature that demonstrates an insufficient number of sturdy detection schemes that can recognize the small semantic aberrations inherent in LLM-generated deceptive text. Our proposed co-design hybrid model is Semantic Anomaly Detection with Isolation Forest (SADI) model that combines the synergistic mixture of a fine-tuned transformer-based LLM for deep semantic feature extraction with Isolation Forest algorithm that detects anomalies efficiently. This study introduces SADI, an adaptive semantic-anomaly detector for large-language-model phishing emails. Using a corpus of 10 000 messages, SADI attains an F1 score of 0.981 (95 % CI 0.978–0.984) and processes a …

Data Privacy Regulations In The Gaming Industry: A Comparative Analysis Of Singapore, Macau, And Japan, Miloslava Plachkinova

Faculty Articles

This study explores the relatively under-researched area of comparing data privacy regulations and best practices across different countries, with a focus on the gaming industry. It provides an overview of general data privacy principles and existing global regulations, analyzing how gaming operators leverage personal data for competitive advantage. Specifically, the research examines the data privacy approaches and regulatory requirements in Singapore, Macau, and Japan, highlighting the cultural and historical contexts influencing these regulations. Through a comparative analysis, the article discusses the compliance needs for gaming operators in these jurisdictions.

Knowledge And Ontology Enhanced Approach To Natural Language Understanding (Koe-Nlu) In Computational Social Media And Healthcare, Naga Usha Gayathri Lokala

Theses and Dissertations

Natural Language Understanding (NLU) faces both opportunities and challenges as the amount of social media and healthcare data grows. This is particularly evident in context-sensitive applications such as evaluating cognitive health, identifying mental health symptoms, and monitoring drug abuse. Even though traditional NLU models work well for processing language in a wide range of areas, they often lack the ability to understand language in a specific domain, reason in context, and incorporate structured external knowledge. This dissertation talks about the Knowledge and Ontology Enhanced Approach to Natural Language Understanding (KOE-NLU), a new framework that is meant to make NLU systems …

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 …

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 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 …