Digital Commons Network^™

Achieving More Accurate And Interpretable Fraud Detection With Double Machine Learning, Jeremy Andrew Berry

Theses and Dissertations

Fraud detection remains a critical challenge across industries such as insurance, healthcare, finance, and government. Global losses from fraud and financial crime are estimated in the trillions annually, including billions in healthcare and insurance fraud alone. While effective for prediction, traditional machine learning methods often lack causal interpretability and struggle to adapt to evolving fraud tactics. This dissertation investigates the application of Double Machine Learning (DML), an emerging causal inference technique, to enhance both the accuracy and interpretability of fraud analytics. The research compares DML against established causal inference approaches, leveraging a meta-learning framework to evaluate model performance on accuracy, …

Patterns Of Llm Weaponization: A Comparative Analysis Of Exploitation Incidents Across Commercial Ai Systems, George Antoniou

Faculty and Staff Publications & Presentations

This comparative study examines patterns of Large Language Model (LLM) weaponization through systematic analysis of four major exploitation incidents spanning 2023-2025. While existing research focuses on isolated incidents or theoretical vulnerabilities, this study provides the first comprehensive comparative framework analyzing exploitation patterns across state-sponsored cyber-espionage (Anthropic Claude incident), academic security research (GPT-4 autonomous privilege escalation), social engineering platforms (SpearBot phishing framework), and underground criminal commoditization (WormGPT/FraudGPT ecosystem). Through comparative analysis across eight dimensions—adversary sophistication, target selection, exploitation techniques, autonomy levels, detection evasion, attribution challenges, defensive gaps, and capability democratization—this research identifies critical cross-case patterns informing defensive prioritization. Findings reveal three …

Ankimedbench: Evaluating Hierarchical Medical Knowledge In Language Model Embeddings, Neel Patel

UNLV Theses, Dissertations, Professional Papers, and Capstones

Despite achieving over 90% accuracy on medical benchmarks, recent studies show physicians cannot effectively leverage language models to improve clinical reasoning. Current benchmarks test isolated factual recall, but clinical practice requires hierarchical navigation through diagnostic categories—starting broad and narrowing systematically from chest pain to cardiovascular pathology to myocardial infarction to specific STEMI types. Existing evaluations cannot measure whether models preserve this taxonomic structure essential for clinical reasoning.

We introduce AnkiMedBench, built from 16,512 medical flashcards used by students preparing for licensing exams. Cards are organized across six hierarchy levels spanning 16 broad medical specialties to 672 specific diseases and conditions. …

Polyminhash: Efficient Area-Based Minhashing Of Polygons For Approximate Nearest Neighbor Search, Alima Subedi, Sankalpa Pokharel, Satish Puri

Computer Science Faculty Research & Creative Works

Similarity searches are a critical task in data mining. As datasets grow larger, exact nearest neighbor searches quickly become unfeasible, leading to the adoption of approximate nearest neighbor (ANN) searches. ANN has been studied for text data, images, and trajectories. However, there has been little effort to develop ANN systems for polygons in spatial database systems and geographic information systems. We present PolyMinHash, a system for approximate polygon similarity search that adapts MinHashing into a novel 2D polygon-hashing scheme to generate short, similarity-preserving signatures of input polygons. Minhash is generated by counting the number of randomly sampled points needed before …

A Trigger For The Autonomous Decommissioning Of Smart Devices, Ravindra Mangar, Jared Chandler, Jingyu Qian, Carl A. Gunter, Timothy J. Pierson, David Kotz

Dartmouth Scholarship

Smart devices are ubiquitous in modern environments, yet their decommissioning phase remains poorly studied and often overlooked in system design. We define secure decommissioning as the process by which a smart device securely disconnects from its environment and makes sensitive data inaccessible. If not decommissioned, devices may retain sensitive information – such as security credentials or user-behavior data that could be recovered by an adversary. Unfortunately, some users may forget to decommission a device when they dispose or sell it, and cannot decommission a device that is lost or stolen. This paper investigates a trigger mechanism for individual wireless smart …

Developing An Ai-Assisted Grading System Using Large Language Models, Andrei Modiga

MS in Computer Science Project Reports

We present a grading system that accelerates evaluation of open-ended student work across scanned and digital workflows. The system crops answer regions from PDFs, assigns submissions via OCR on identity regions only, and groups answers by visual semantics using a vision LLM. Instructors review and edit groups, apply rubric items once per group, and export grades from an on-screen table. The solution integrates Ghostscript rasterization, PdfPig page orchestration, SkiaSharp region extraction, Tesseract identity OCR, and GPT-4o Vision for grouping. We detail the architecture, token-budgeted batching strategy, and persistence design, then describe testing results for grouping quality, time-on-task, and usability. The …

Modern Technology Addiction: Developer Duty Of Care, Jonah Hampton

Honors College Theses

Technology addiction includes any frequent use of technology which interferes in the user’s life. The subject continues growth as an epidemic and research field, yet prior literature does not often analyze the role of technology developers. This study performs a literature and legal synthesis to evaluate user and company responsibility, implications of responsibility, and promising solutions. Post 2020 literature was selected for coverage on context, addictive features, effects, solutions, or perspectives on law. Legal examples from different addiction industries were also selected for analysis to understand previous precedents. The study found a pattern of addictive traits, persuasive design, and recurring …

Flooding Behavior Near The Us/Canada Border: Complications And Approaches, Maria T. Dodson

Honors College Theses

Flood forecasting remains a major challenge due to the nonlinear nature of hydrological systems and uncertainties in environmental data. This study aimed to address the prevalent challenges that arise from forecasting flooding behavior. To address the inherent complexity of hydrological forecasting, a machine learning framework was developed and trained on major contributing factors. To achieve an optimal balance between computational efficiency and predictive performance, a Gated Recurrent Unit (GRU) was selected as the optimal machine learning model. As the chosen dataset, North American Land Data Assimilation System Phase 2 (NLDAS2), is known to have inaccuracies in the important feature Relative …

Application Of Augmented Reality Technology As A Dietary Monitoring And Control Measure Among Adults: A Systematic Review, Gabrielle Victoria Gonzalez, Bingjing Mao, Ruxin Wang, Wen Liu, Chen Wang, Tung Sung Tseng

School of Public Health Faculty Publications

Background/Objectives: Traditional dietary monitoring methods such as 24 h recalls rely on self-report, leading to recall bias and underreporting. Similarly, dietary control approaches, including portion control and calorie restriction, depend on user accuracy and consistency. Augmented reality (AR) offers a promising alternative for improving dietary monitoring and control by enhancing engagement, feedback accuracy, and user learning. This systematic review aimed to examine how AR technologies are implemented to support dietary monitoring and control and to evaluate their usability and effectiveness among adults. Methods: A systematic search of PubMed, CINAHL, and Embase identified studies published between 2000 and 2025 that evaluated …

Toward A Unified Network Flow Framework: From Conservation Principles To Fluid Dynamics Models, Zonghan Zhang

Theses and Dissertations

Network flows govern a wide range of critical systems, from tangible infrastructures like transportation and power grids to replicable processes such as information spread and epidemics. While tangible flows obey conservation laws and physical constraints, replicable flows, like rumors or viruses, can grow, decay, or vanish unpredictably. Despite their increasing interaction in real-world settings, these flow types are typically modeled in isolation, using disconnected mathematical frameworks. This dissertation presents a unified modeling approach that bridges the gap between conserved and replicable flows by embedding principles from fluid dynamics into graph-based propagation models. I introduce physically informed extensions to classical models, …

Using Quantum Annealing For Sampling And Pattern Generation In Generative Machine Learning And Catastrophic Forgetting Mitigation, Abdelmoula El Yazizi

Theses and Dissertations

The first goal of this dissertation was to understand the reasons for the absence in previous investigations of significant and consistent improvements in the trainability of Restricted Boltzmann Machines (RBM) when the Quantum Annealer (QA) was used for sampling from the RBM probability distribution. The second goal was to address the shortcomings of those previous investigations, explore possibilities of improving RBM training, and identify other machine learning applications that could benefit from sampling or from generating patterns by the QA. The first part of this dissertation focused on a Local-Valley (LV) centered approach to assessing the quality of sampling. QA-based …

Impacts Of Upgrading Gcc On The Srtuner Compiler Autotuning Algorithm, Jonathan D. Butterfield

Theses and Dissertations

An often-overlooked research gap is the applicability of research over time. The impact of updating the GNU Compiler Collection version was investigated for the SRTuner compiler optimization research. Versions 7 through 14 were tested using documented flags and recommended flag dependencies. Quality controls measured and enforced consistency across test runs and benchmarks. The performance of SRTuner was compared to a random control while the impact of enforcing flag dependencies was evaluated. The most influential flags were investigated by turning them off in isolation, but results were mixed. It was determined that enforcing flag dependencies resulted in improved performance, but the …

Causal Structure Discovery For Explaining Important Features In Machine Learning, Yina Hou

Tennessee State University Alumni Theses and Dissertations

Understanding causal relationships between input variables and outcomes is critical for scientific advancement. The famous adage in medicine, "correlation does not imply causation," holds profound significance for understanding disease etiology. Although explainable machine learning (ML) has shown major advances in predictive modeling, it remains unclear how ML-derived important variables relate to causal variables. This thesis investigates the association between causal variables and those identified as correlated and important for ML-based prediction to bridge a critical knowledge gap in data science. Specifically, the thesis explores two research questions: when and to what extent (1) statistically correlated variables are also causal? and …

Cross-Domain Transfer Learning Of Tabular Data With Heterogeneous Feature Spaces, Kazi Fuad Bin Akhter

Tennessee State University Alumni Theses and Dissertations

Transfer learning is a cornerstone of artificial intelligence (AI), enabling the learning of new domains with limited labeled data by reusing knowledge from large-scale foundation models. Transfer learning has achieved remarkable success with vision and language models due to the homogeneity in image and text data. In contrast, a heterogeneous feature space, structured in rows and columns as tabular data, remains underexplored in the transfer-learning literature. The inductive bias of tabular data with heterogeneous feature spaces from disparate application domains is very different from image and text data, which complicates transfer learning. Several recent studies have attempted within-domain and limited …

A Fault Tolerant Honeybee Behaviour Load Balancing Algorithm, Charmaine Adegbite

Tennessee State University Alumni Theses and Dissertations

Cloud computing enables flexible, scalable access to virtualised infrastructure but remains challenged by maintaining performance in fault-prone environments. Faults such as virtual machine failures can degrade task allocation and system reliability. However, traditional load balancing algorithms and bio-inspired strategies, including the Honeybee Behaviour Load Balancing algorithm, assume fault-free conditions. This thesis introduces a Fault-Tolerant Honeybee Behaviour Load Balancing algorithm that enhances the original Honeybee Behaviour Load Balancing algorithm by integrating fault detection. The proposed algorithm monitors the health of VMs and dynamically reallocates workloads when faults are detected which improves stability and reliability. The proposed fault tolerant algorithm was implemented …

Parameter Informed Reinforcement Learning For Vehicle System Identification, Nathan Schaff

Doctoral Dissertations and Master's Theses

Accurate system identification is essential for modeling and controlling vehicle dynamics. This dissertation explores the application of Parameter Informed Reinforcement Learning (PIRL) as a novel approach to system identification (SYSID). PIRL integrates prior system knowledge, such as physical parameters, into reinforcement learning (RL) frameworks to improve estimation accuracy. The study begins with an overview of traditional SYSID methods and then introduces PIRL as a modification of standard RL. The research applies PIRL to short-period aircraft dynamics, demonstrating its effectiveness in both offline and online learning frameworks. The dissertation then further explores PIRL’s utility in an indirect model reference adaptive control …

Challenges In Engineering Machine Learning (Software) Systems, Raffi T. Khatchadourian Ph.D.

Open Educational Resources

Lecture slides on the software engineering challenges unique to machine learning systems, for an undergraduate software engineering course. After contrasting traditional programming with machine learning, the deck examines why the usual tools for managing complexity—abstraction, reuse, and composition—are harder to apply to ML, given the lack of clear specifications and modularity. It covers concept drift, feedback loops (illustrated with crime-prediction and recommendation examples), and the accumulation of technical debt in ML systems, including the role and pitfalls of notebooks in moving from experimentation to production. Based on "Machine Learning in Production/AI Engineering" by Christian Kaestner and Eunsuk Kang (Carnegie Mellon …

My First Conversation With Chatgpt (February 22, 2023): Origins Of A Generative Dialogue, David Smith

Publications and Research

This working paper presents the first recorded interaction between the author and the generative AI system ChatGPT, written on February 22, 2023 during the initial weeks of a faculty sabbatical in Boston. The document preserves a complete and unedited transcript of an exploratory conversation conducted without predetermined research aims, marking the author’s first encounter with a large-language-model conversational interface. Although the exchange includes creative experimentation—including musical and poetic prompts—the discussion remains informal and wide-ranging, and no theoretical framework is articulated at this stage. Rather, this transcript is published as primary-source material documenting the moment of discovery and experimentation that precedes …

Implementing Dataops: A Scalable Framework For Modern Data Warehousing, Dmytro Valiaiev

Theses and Dissertations

DataOps has been coined as a novel term that emerged as a synthesis of data management practices with software engineering concepts, such as DevOps and Agile, with the goal of improving data quality and governance in enterprises. The proliferation of scratch table use and transformation tools, such as dbt, has led to an exponential increase in the number of data models, which complicates standardization efforts and increases maintenance overhead. Although the market is saturated with various flavors of text-to-SQL engines that promote increased productivity and self-service use in organizations, there are limited tools available to optimize individual queries, enforce consistency, …

Towards Fair Sequential Resource Allocation: Algorithmic Designs, Interventions, And Evaluations, Ashwin Kumar

McKelvey School of Engineering Graduate Student Theses & Dissertations

This thesis develops a comprehensive framework for fair sequential resource allocation in multi-agent systems where a centralized allocator coordinates actions under global feasibility constraints, while satisfying preferences of different agents. Ranging from ridesharing platforms and homelessness intervention programs to power grid management, such systems play a critical role in shaping access to essential resources. Yet, existing approaches to resource allocation often prioritize aggregate utility, leading to systematic inequities across individuals and groups, particularly in sequential settings where decisions unfold over time. To address this challenge, we introduce the Distributed Evaluation, Centralized Allocation (DECA) framework, which unifies a broad class of …

Online Decision Mamba, Trenton W. Ruf

Dissertations and Theses

Online in-context reinforcement learning enhances offline-trained policies through online fine-tuning. We introduce Online Decision Mamba (ODM), an architecture that replaces the attention mechanism in Online Decision Transformers (ODT) with the Mamba module to improve long-context sequence modeling and overall RL performance. We performed in-depth evaluations on MuJoCo (OpenAI Gym) and Atari benchmarks, comparing ODM against state-of-the-art offline and online baselines—including Decision Mamba (DM) and ODT. Our results show that ODM achieves competitive or superior performance, with particularly robust gains when initial datasets lack expert demonstrations. In the Qbert Atari environment, ODM shows context-length sensitivity similar to offline DM; however, we …

Prism (Proxy Recognition And Inclusion Scoring Method), Destiny Raburnel, Crystal Tubbs, Md Abdullah Al Hafiz Khan

Symposium of Student Scholars

AI-driven automated hiring tools are reshaping how companies find talent, but they often reproduce the hidden biases embedded in their training data. Our project, PRISM (Proxy Recognition and Inclusion Scoring Method), investigates how subtle demographic signals, specifically first names associated with gender and race, influence AI resume screening even when candidates have identical qualifications. We built a controlled dataset of resumes that are identical in every way except for the applicant's first name, with each resume using a racially neutral surname to isolate how first names alone affect scoring. We tested these resumes against job postings in technology, healthcare, and …

Career: End-To-End Active Region-Based Heliospheric Forecasting System Using Multi-Spacecraft Data And Machine Learning, Soukaina Filali Boubrahimi

Funded Research Records

No abstract provided.

Dynamicslab: Interactive Physics Simulations For Intermediate Classical Mechanics, John M. Edwards, Boyd Farrell Edwards, Hillary L. Swanson

Funded Research Records

No abstract provided.

Shine: Understanding The Relationships Of Photospheric Vector Magnetic Field Parameters In Solar Flare Occurrences Using Graph-Based Machine Learning Models, Shah Hamdi

Funded Research Records

No abstract provided.

How Novices Write Code: Discovering Best Practices And How They Can Be Adopted, John Martin Edwards

Funded Research Records

No abstract provided.

Bcser: Learning Analytics For Process-Driven Computer Programming Assignments, Hamid Karimi

Funded Research Records

No abstract provided.

Predicting Skin Concern Severity From Genetic And Lifestyle Factors: A Comparative Multi-Output Machine Learning Framework, Yassine Benachour, Lina Maloukh, Sadok Bouamama, Barbara Geusens

All Works

Personalized dermatology increasingly leverages both genetic predispositions and lifestyle behaviors to model individual skin health outcomes. This study proposes a multi-output machine learning framework to predict the severity of six dermatological phenotypes—acne, redness, dryness, sensitivity, scarring, and pigmentation—using a multimodal dataset of 5,254 individuals. Input features include mutation profiles for six skin-related genes (FLG, MMP1, MMP3, AQP3, SOD2, GPX) and 22 lifestyle variables such as sun exposure, stress, and hydration. We train and evaluate LightGBM models under independent, multi-output, and chained configurations. Performance is assessed using Mean Absolute Error (MAE) and average Quadratic Weighted Kappa (QWK). The proposed ordinal-aware independent …

Artificial Intelligence In Insurance Fraud Detection: Applications And Implications For Internal Audit, Maria Davis

Honors Projects

Artificial Intelligence (AI) is being used in accounting and the insurance industry to assist with auditing and fraud detection. The Big Four public accounting firms have invested heavily in AI implementation efforts. Across these firms, AI has been used to reallocate auditors’ time from mundane tasks to more complex tasks that require human judgement. Within the insurance industry, machine learning, deep learning, and natural language processing, among other AI tools, have proven helpful in fraud detection efforts. While the positive impacts of AI usage are clear, concerns surrounding the replacement of human jobs, a lack of transparency in auditing, heavy …

Systematics And Systems Theory: Reconstructability Analysis Of The Tetrad, Martin Zwick

Complex Systems Faculty Publications and Presentations

This talk discusses the relationship between systems theory, specifically Reconstructability Analysis, and Systematics, a systems theory-like framework of number symbolism developed by John G. Bennett, which he presented in his four-volume magnum opus, The Dramatic Universe. The talk, given to a community of people interested in Bennett's ideas, focuses on Martin Zwick's paper "Ideas and Graphs: the Tetrad of Activity" archived at https://archives.pdx.edu/ds/psu/36249.