Physical Sciences and Mathematics Commons^™

An Empirical Framework For Evaluating Semantic Preservation Using Hugging Face, Nan Jia, Anita Raja, Raffi Khatchadourian

Publications and Research

As machine learning (ML) becomes an integral part of high-autonomy systems, it is critical to ensure the trustworthiness of learning-enabled software systems (LESS). Yet, the nondeterministic and run-time-defined semantics of ML complicate traditional software refactoring. We define semantic preservation in LESS as the property that optimizations of intelligent components do not alter the system's overall functional behavior. This paper introduces an empirical framework to evaluate semantic preservation in LESS by mining model evolution data from HuggingFace. We extract commit histories, $\textit{Model Cards}$, and performance metrics from a large number of models. To establish baselines, we conducted case studies in three …

Color-Blind Resilience: How Uniform Climate Standards Reproduce Environmental Inequality In The Rockaways, Aaryan M. Nair

Publications and Research

This paper examines how ostensibly uniform climate resilience standards can reproduce environmental inequality in socially uneven landscapes, using the Rockaway Peninsula in New York City as a critical case study. While flood-resistant building codes, zoning regulations, and insurance frameworks are designed to provide equal protection across flood-prone areas, this analysis argues that their “color-blind” application obscures and intensifies underlying disparities in financial capacity, housing conditions, and tenure. Drawing on environmental justice (EJ) theory, the paper identifies two key mechanisms through which inequality is reproduced. First, “code without capacity” demonstrates how uniform technical standards—such as elevation requirements under NYC Building Code …

Resilience For Whom? A Study In Pathways And Production Of Unequal Risk In The Rockaways, Aaryan M. Nair

Publications and Research

This paper investigates how climate adaptation policies in the Rockaway Peninsula of New York City produce unequal outcomes despite their ostensibly neutral design, framing resilience as an environmental and climate justice problem. Situated within a landscape marked by stark socioeconomic disparities, aging housing stock, and high exposure to coastal flooding, the study demonstrates that existing adaptation strategies—centered on uniform technical standards, insurance frameworks, and large-scale coastal infrastructure—systematically advantage actors with greater financial and institutional capacity while leaving low-income residents, particularly renters, disproportionately vulnerable.

The analysis identifies three core dynamics underlying this “unequal-by-design” resilience regime: (1) the stratified feasibility of compliance …

All Games Have Equilibria, M. Ali Khan, Arthur Paul Pedersen, Maxwell B. Stinchcombe

Publications and Research

Research on Nash equilibrium existence for infinite games has grown into a patchwork of technical preconditions and counterexamples. This paper presents a unified program in equilibrium theory by revising the predominant model of mixed strategies based on countable additivity. A game is specified by a nonempty set of players and, for each player, a nonempty action set and a bounded von Neumann-Morgenstern utility function. Every such game is shown to admit a Nash equilibrium in finitely additive mixed strategies. In addition, the equilibrium correspondence for any such game is shown to be nonempty, compact-valued, and upper hemicontinuous, and the same …

Spectral Decimation On The Schreier Graphs Of The Basilica Group: A Thesis In Fractal Analysis, Anne D. Bannon

Scripps Senior Theses

This thesis is intended to provide a comprehensive overview of the literature required to fully understand research conducted during the University of Connecticut's Fractals & Stochastics REU in the summer of 2025. The literature review includes a description of Robert Strichartz's seminal work pertaining to the Laplacian spectrum of the Sierpiński Gasket, which provides a framework for how we approach studying the spectrum of the basilica Julia set. Defining the basilica Julia set and the closely-related Basilica group involves graph theory, automata theory, iterated monodromy group theory, and amenable group theory. Further time is dedicated to defining the graph Laplacian …

Modeling French Language Preservation: An Optimal Control Problem, Charlotte Blasi

Scripps Senior Theses

The French language is one of the most globally spoken languages, with over 300 million speakers and designated the official language of 29 nations. French continues to prosper as a result of its titular nations' imperialist history that emphasized linguistic diffusion as well as the support of international organizations dedicated to promoting both the language and the culture of French-speaking nations. In this thesis, we explore the nuanced history of French-speaking countries and one institution dedicated  to promoting the French language, the OIF. We further examine the media mechanism of the OIF and construct a system of Ordinary Differential Equations …

An Introduction To Modern Conversations On Knot Invariants, Stella Shah

Scripps Senior Theses

Knot Theory is a vast and diverse subfield of modern mathematics involving the classification and abstraction of knots and links. In this thesis, we wish to provide the necessary background for and an explanation of two papers in different subfields of knot theory, The Forbidden Quiver of a Link, and Biquandle Fares and Link Invariants.

In Chapter I, we begin with an introduction to knot theory and knot invariants. We continue to present the example of Fox Colorings, and conclude the chapter with an example of the Fox Coloring Number Invariant.

In Chapter II, we explore the derivation and utilization …

Large Aerosol Particles Favor Haze Conditions Through Limitations On Water Budget And Activation Kinetics, Hamed F. Sadi, Ernie Lewis, Kadja Flore Gali, Fan Yang, Jesse Anderson, Suryadev Pratap Singh, Jae Min Yeom, Will Cantrell, Raymond Shaw

Michigan Tech Publications

Experiments in the Pi Convection‐Cloud Chamber conducted by systematically changing the diameter of injected dry aerosol particles while holding the temperature difference constant demonstrate that dry diameter strongly influences the onset of haze‐dominated conditions. Two factors contribute: the system becomes water‐limited, resulting in reduction of supersaturation by growing aerosol particles to the activation diameter; and the activation process becomes kinetically limited. Dry aerosol diameter exerts a strong influence on activation time, with a power‐law exponent of 9/2. Kinetically limited activation occurs when the ratio of the activation and droplet residence times is greater than unity. The findings demonstrate that a …

Doikayt And The Ethics Of Belonging: Lessons In Place-Making From The Jewish Labor Bund, Zoë L. Jacobs

Scripps Senior Theses

This thesis traces the evolution of doikayt—the Bundist principle of hereness—from its origins in the Jewish Labor Bund to contemporary applications in North America. Rooted in a refusal of both assimilation and nationalist escape, doikayt emphasizes cultivating safety, solidarity, and belonging through collective responsibility rather than territorial claims.

Building on this foundation, the exploration situates hereness within a settler-colonial context, where ethical dwelling is complicated by Indigenous dispossession. Drawing on Alexis Shotwell alongside Indigenous thinkers Vanessa Watts and Kyle Whyte, I examine how inhabiting place requires confronting historical injustices and ongoing relational responsibilities. Chapter 3 analyzes Linke Fligl (2016–2022) as …

Polysemicolon; Novice Programmers And Java Keywords, Briana C. Bettin

Michigan Tech Publications

Most industrial programming languages leverage the English language for reserved keywords – words which a program compiler recognizes as specific execution commands. The divide between expert and novice programmers showcases an intriguing middle-ground by which the polysemy of many keywords becomes revealed. This essay explores a sampling of the multitude of keyword interpretations that a novice programmer may derive from the Java language’s syntactic style and keywords specifically, and how the polysemy of both “English” and “code” meanings to these terms affects the novice-expert programmer transition.The transition from novice to expert, and the mapping of the career of metaphor to …

Real Interpolation: An Approximate Introduction, Madeline Anderson

Scripps Senior Theses

This thesis provides an introduction to real interpolation. We establish

relevant notions in functional analysis first, and use these concepts to study

real interpolation using J. Peetre’s 𝐾-functional in some detail. We also

explore the basics of approximation theory, in particular the connection

between approximation and interpolation results.

Understanding Phishing Susceptibility Through Expert Consensus Using Digital Marketing Parallels And A Machine Learning-Based Implementation, Mansoor Ahmad

All Graduate Theses, Dissertations, and Other Capstone Projects

Phishing remains one of the most effective attack vectors for gaining unauthorized access to organizational systems, yet defenders often lack systematic methods to assess their exposure before an attack. This study develops a framework that uses machine learning to encode the collective expertise of cybersecurity practitioners into a portable phishing susceptibility assessment tool, with the goal to help security teams proactively identify patterns, prioritize awareness training, and strengthen detection controls. The study surveyed 27 practitioners with extensive experience in social engineering, red teaming, penetration testing, and threat analysis to identify which factors most influence phishing susceptibility. Practitioners provided quantitative ratings …

Impacts Of Management Practices And Soil Properties On Free-Living Nitrogen Fixation In Eastern South Dakota, Laura Galvan Nuevo

Electronic Theses and Dissertations

Free-living nitrogen (N) fixation (FLNF) may represent a sustainable pathway to supplement crop N demand in agroecosystems, yet edaphic and management factors regulating this process under field conditions remain poorly understood in the Northern Great Plains. Using 15N2 direct method to quantify potential FLNF (pFLNF) and baseline FLNF (bFLNF), this thesis investigates how crop rotations (2-, 3-, and 4-year systems), tillage management (conventional vs. no-till), cover cropping, and soil properties affect the potential of free-living diazotrophs to supply N via FLNF in a long-term no-till corn-based system in Southeastern South Dakota. A two-year field study evaluated pFLNF across four growing …

Ai Vs. Genai: Combating Llm Generated Prescription Fraud With Transformer Based Detection Models, Ankitha Vokkaleri Shankarappa

Selected Full-Text Master Theses 2021-

The rapid evolution of Large Language Models (LLMs) has introduced a sophisticated new vector for healthcare fraud: the generation of high-fidelity, synthetic medical prescriptions. Traditional fraud detection systems, which rely on rule-based engines and basic statistical anomalies, are increasingly ill-equipped to identify these AI-generated forgeries that mimic the structural and clinical logic of authentic records. This thesis presents a robust detection framework using Transformer-based architectures to distinguish between human-authored Medicare Part D prescriptions and fully synthetic records generated by GPT-4.

The research was conducted across two distinct phases: an initial pilot study using 4,000 samples and a rigorous validation stress …

Erosion Of Trust In Online Information, Tirth Desai

A with Honors Projects

Researching how AI spreads misinformation and impacts trust in information.

Collaboration And Co-Management Ahead Of Permitting: Understanding How Actors And Their Interactions Lead To Non-Optimal Shoreline Projects, Juita-Elena (Wie) Yusuf, Mariana Saitgalina, Michelle Covi

School of Public Service Faculty Publications

Living shorelines are widely promoted as nature-based solutions to coastal erosion and wetland protection, yet hardened shoreline structures continue to dominate even in jurisdictions with explicit policy mandates prioritizing living shorelines. In this research, we examine why non-optimal shoreline modification outcomes persist in Virginia (USA) despite a regulatory framework designed to promote ecological alternatives. We use primary data from interviews with wetlands board members, marine contractors, and nonprofit organizations and findings from a secondary survey of shoreline property owners to analyze shoreline management as a multi-sector collaborative decision-making process. Findings show that shoreline outcomes are shaped less by individual regulatory …

Fixed Dune Grassland Mapping, Ronica Reyes, Madison Mayenschein, Jared Cruz

Environmental Science & Management Senior Capstones

Our  study applies Geographic Information Systems (GIS) and field-based data collection to map and assess a degraded fixed dune grassland parcel in Humboldt County, CA. Our project supports restoration planning led by Friends of the Dunes by identifying site features, habitat composition, and disturbance patterns at the site. Using GPS surveys, aerial imagery, and ArcGIS Pro, we mapped infrastructure, cement footprints, debris piles, vegetation types, and habitat boundaries. We mapped infrastructure such as concrete foundations and debris, which contribute to ongoing ecological degradation and the presence of invasive species. In addition, our spatial analysis highlights variability in disturbance. From this …

Healing Our Life Source: Riparian Eco-Cultural Restoration In Blue Lake Rancheria At School Creek, A Tributary Of Baduwa't In Northern California, Rae Mcgrath, Lily Moore, Keneti Jesus Pinones, Colin Rogers, Aberdeen Spade

Environmental Science & Management Senior Capstones

School Creek, a tributary in the Baduwa’t (mad river) Watershed, is a riparian corridor that is key habitat for Coho Salmon. This project is under the management of the Blue Lake Rancheria, located on the ancestral home of the Wiyot people, and was conducted by cal poly humboldt students. The restoration of School Creek is an ongoing project aimed at repairing the health of the riparian corridor, improving the creek’s channel structure, and creating a diverse community of culturally significant plant species. Previous restoration efforts have established plantings to further develop the riparian habitat and are surveyed within this report. …

Removal And Monitoring Of Ammophila Arenaria In A Previously Restored Foredune Site In Humboldt Coastal Nature Center, Lucas Benjamin Griffin, Haylee Lynn Kimball, Maya Contreras, Ava Christenson

Environmental Science & Management Senior Capstones

Coastal dunes are important ecosystems that provide vital habitat for rare and endangered species as well as vital ecosystem services. However, since the late nineteenth century, coastal dune ecosystems have been invaded and altered by Ammophila arenaria, which is a beach grass species native to Europe. The dunes within the Humboldt Coastal Nature Center are one such example of this. This property is currently managed by the non-profit organization Friends of the Dunes, which has been working to restore the dunes' ecology and topography. Our project focuses on the monitoring and maintenance of a 2021 restoration project within a 17,794 …

Vegetation Management Plan And Wildlife Enhancement Project For The Woodley Island Wildlife Area, Sophie Belle Shapero, Brandon Carl Bauch, Moises Ortiz, Angelina Espinoza

Environmental Science & Management Senior Capstones

The Woodley Island Vegetation Management Plan and Wildlife Enhancement Project was created by Cal Poly Humboldt students in partnership with the Humboldt Bay Harbor Recreation and Conservation District as a capstone senior project. This project provides a basic management plan for invasive species and offers suggestions for how to enhance the wildlife areas on the island. The goals of the project are to provide maps of the habitat types within the wildlife reserves, map areas of concern to be targeted for invasive species removal, and provide guidelines for how to increase habitat diversity, wildlife forage, and decrease invasive species presence. …

Woody Ryno Farms Native Prairie Plant Restoration & Invasive Management Conservation Management Report, Jezebel Kenny, Rowan Haeger, Ian Mendonca, Annika Weihl

Environmental Science & Management Senior Capstones

Coastal prairies are biodiverse and sensitive landscapes that support a variety of grassland species in mediterranean climates. Hedgerows may act as a natural barrier for coastal prairies or agricultural regions, aiding in natural protection and support for biodiversity. Humboldt County, California, hosts a number of agricultural regions, including Woody Ryno Farms, a small-scale agricultural operation, on the edge of Dows Prairie near the coast of Mckinleyville. This project aims to continue restoration activity being conducted on the site, helping to increase native biodiversity and agricultural productivity. Project objectives include conducting a plant assessment, applying treatments to individual plants where necessary, …

A Comprehensive Survey On Facial Expression Generation: From Gans To Llm-Guided Multimodal Models, Murad Hasan, Rabab Abdelfattah, Kareem Abdelfatah, Mostafa Fouda, Ahmed Sherif

Faculty Publications

Facial expression generation (FEG) has emerged as a vital area in human–computer interaction, virtual avatars, and affective computing, aiming to synthesize natural and expressive facial behaviors across diverse interaction contexts. This survey presents a comprehensive analysis of recent advances in FEG, organized into six key paradigms: speech-driven expression generation, facial reaction generation, face video generation, facial animation, avatar-based generation, and text-driven expression generation. We review a wide range of model architectures, including VQ-VAEs, Generative Adversarial Networks (GANs), 3D Morphable Models (3DMMs), Transformers, and diffusion-based approaches, and compare their performance using commonly adopted evaluation metrics such as Fréchet Distance (FD), Peak …

Body Mass Index Has No Impact On Complications And Mortality For Patients With Stage Iv Pancreatic Ductal Adenocarcinoma, Ren Bryant, Hannah Darnell, Megan Hall, Kelsey N. Karnik, Kristen J. Mcquerry, Ruben R. Plentz

Biostatistics Faculty Publications

Background: Pancreatic ductal adenocarcinoma (PDAC) is one of the leading causes of United States (USA) cancer death. Overweight and obesity developing into a growing global medical and socio-economic problem, affecting approximately 42% of adults in the USA population. The aim of our analysis was to evaluate the influence of overweight and obesity on complications and clinical outcome in patients with stage IV PDAC.

Methods: We retrospectively reviewed electronic health records of patients diagnosed with stage IV PDAC (n=162) who followed with the University of Kentucky from January 2017–October 2024. Comparisons were based on the body mass index (BMI): low BMI …

Flood Risk Prediction System For Irish River Networks, Fernando Adalberto Naatz Heringer, Hugo Rodrigues De Freitas

ICT

Flooding costs the Irish economy more than 300 million euros annually; however, the hydrometric and meteorological data needed to predict dangerous river rises is already collected and freely available. This project applies machine learning to a decade of hourly observations from OPW Station 09001 on the River Liffey at Leixlip and Met Eireann Casement Aerodrome to build a flood risk prediction system with a 72-hour forecast horizon.

A two-phase modelling pipeline was built following the CRISP-DM framework. Phase 1 compared LSTM and GRU recurrent neural networks on the task of forecasting the river level in metres 72 hours ahead. LSTM …

Machine Learning System For Cereal Yield Prediction In South America., Victor Gabriel Oliveira, Kelvin Henrique Ferreira Dumas

ICT

This project developed a machine learning solution to predict cereal yield in South America using FAOSTAT agricultural data. The objective was to estimate yield in kg/ha for rice, maize, and wheat based on variables such as country, crop type, year, harvested area, producer price, pesticide use, and nutrient indicators.

The project followed the CRISP-DM methodology, covering business understanding, data understanding, data preparation, modelling, evaluation, and deployment. Several regression models were tested, including Random Forest, Gradient Boosting, AdaBoost, XGBoost, KNN, and SVR.

The models were evaluated using MAE, MSE, RMSE, and R² score. The results showed that tree-based ensemble models performed …

A Comparative Analysis Of Supervised Machine Learning Techniques For Predicting It Incident Resolution Time, Ammad Hussain, Hussnain Yaqoob

ICT

IT support teams are facing the persistent challenge in estimating how long an incident will take to resolve. The lack of reliable predictions, planning decisions around staffing, escalation and user communication are largely guesswork. This project set out to address that gap by building a machine learning solution capable of predicting incident resolution time from details available at the point a ticket is logged.

Using the UCI IT Incident Event Log dataset, which contains 141,712 records from a real service management system, the project followed the CRISP-DM framework across six phases from business understanding through to deployment. After thorough data …

A Comprehensive Study Of Sarima And Xgboost Models For Short-Term Traffic Volume Forecasting At A Melbourne Intersection (Victoria), Australia, Douglas Vinicius Dierings, Felipe Fontanive Marques

ICT

Traffic congestion is an increasing challenge in modern cities, impacting transportation efficiency and urban sustainability. This project focuses on short-term traffic volume forecasting, using 15-minute Sydney Coordinated Adaptive Traffic System (SCATS) data collected from the State of Victoria, Australia. The study compares two forecasting approaches: SARIMA, a statistical time-series model, and XGBoost, a decision-tree machine-learning model. Historical traffic data from 2022 to 2024 was analysed to identify traffic patterns, seasonal trends, and peak traffic periods. The models were evaluated for their effectiveness in handling urban traffic behaviour. The results of this study aim to support smarter traffic management, infrastructure planning, …

Identifying New Programme Development Viability, Grant Goodwin

ICT

Higher Education Institutions (HEIs) face the challenge of developing programmes that respond to changing labour market needs while providing graduates with relevant employment opportunities and ensuring institutional financial viability. Rapid technological and economic changes make it increasingly important for HEIs to use reliable data to identify emerging skills and workforce demands. This project explores the use of machine learning and nationally collected employment data to support evidence-based decision-making in higher education programme development. By analysing patterns and relationships within labour market data, machine learning models can help identify areas of growing demand and provide insights into the potential alignment between …

Seasonal Dublin Airport Passenger Arrivals Forecast For The Economy Sector Of Dublin Airport, Roberto Tavares De Oliveira Neto

ICT

Dublin Airport plays a significant role in Ireland’s tourism and commercial sectors, with passenger demand directly influencing retail, hospitality, car rental and other airport-related businesses. Accurate forecasting of passenger numbers can support these sectors in planning marketing activities, managing stock levels and making evidence-based operational decisions. This project develops a monthly forecasting framework for passenger arrivals at Dublin Airport using data from the Central Statistics Office (CSO). Building on previous exploratory data analysis, the study evaluates five time series forecasting approaches: ARIMA, SARIMA, Holt-Winters Triple Exponential Smoothing, TBATS, and an automated model comparison pipeline using PyCaret. The models are assessed …

Machine Learning For Early Obesity Risk Prediction, Giuseppe Cutugno

ICT

Obesity is a major global public health concern, with significant implications for individual health, healthcare systems, and wider economic outcomes. Early identification of obesity risk can support preventive strategies and encourage healthier lifestyle choices before health conditions become more severe. This project develops a machine learning framework for estimating obesity levels using data relating to individuals’ physical characteristics, eating habits, lifestyle behaviours, and demographic factors. Following the CRISP-DM framework, the project examines the structure and limitations of the dataset, including its mixture of objective measurements, survey-based variables, and synthetically generated observations. Two modelling approaches are considered, one including BMI as …