Physical Sciences and Mathematics Commons^™

Integrating Human Expert Knowledge With Openai And Chatgpt: A Secure And Privacy-Enabled Knowledge Acquisition Approach, Ben Phillips

College of Engineering Summer Undergraduate Research Program

Advanced Large Language Models (LLMs) struggle to produce accurate results and preserve user privacy for use cases involving domain-specific knowledge. A privacy-preserving approach for leveraging LLM capabilities on domain-specific knowledge could greatly expand the use cases of LLMs in a variety of disciplines and industries. This project explores a method for acquiring domain-specific knowledge for use with GPT3 while protecting sensitive user information with ML-based text-sanitization.

Development Of User Interface And Testing Harness, Jacob Amezquita, William Albertini

College of Engineering Summer Undergraduate Research Program

No abstract provided.

Ai For Search And Rescue - Locating A Missing Person, David Hernandez, Sai Rama Balakrishnan, Timmy Chin, Aditya Manikonda, Vasanth Pugalenthi

College of Engineering Summer Undergraduate Research Program

Building on the work done initially as a SURP 2021 project and continued through 2021-23, the focus for this summer project will be on the use of computer technology for locating a missing person. Over the last year, we developed the digital equivalents of about 30 paper-based S&R forms and the infrastructure to collect the respective information. In their current use, these paper forms are filled out by search teams, collected in a command post, and reviewed by search coordinators. This process is time-consuming, prone to errors and loss of information, and relies heavily on the experience, skills, and mental …

Signatures Of Feedback In The Spectacular Extended Emission Region Of Ngc 5972, Thomas Harvey, W. Peter Maksym, William Keel, Michael Koss, Vardha N. Bennert, S. Drew Chojnowski, Ezequiel Treister, Carolina Finlez, Chris J. Lintott, Alexei Moiseev, Brooke D. Simmons, Lia F. Sartori, Megan Urry

Physics

We present Chandra X-ray Observatory observations and Space Telescope Imaging Spectrograph spectra of NGC 5972, one of the 19 ‘Voorwerpjes’ galaxies. This galaxy contains an extended emission-line region (EELR) and an arcsecond scale nuclear bubble. NGC 5972 is a faded active galactic nucleus (AGN), with EELR luminosity suggesting a 2.1 dex decrease in L_bol in the last ∼5 ×10⁴ yr. We investigate the role of AGN feedback in exciting the EELR and bubble given the long-term variability and potential accretion state changes. We detect broad-band (0.3–8 keV) X-ray emission in the near-nuclear regions, coincident with the [O III …

Ethics And Social Justice For Ai In Data Science, Arya Ramchander, Kylene Nicole Landenberger

College of Engineering Summer Undergraduate Research Program

The advances of AI raise several critical questions about human values and ethics, highlighting the need for researchers and developers to consider the ethical implications and the risks of neglecting them. In the past few years, student researchers have developed an AI model that allows users to test their surveys for possible breaches of subject confidentiality. This allows the users to gauge the ethicality of their proposal. This summer, we have expanded on this research and launched an interactive model for students and researches to assess their current work for ethical and social justice implications. Using Langchain and Figma, we …

Exploring Approaches To Engage K-12 Students In Learning Computational Thinking Using Collaborative Robots, Zoila Anuri Kanu

College of Engineering Summer Undergraduate Research Program

Minority students are largely underrepresented in the STEM field. The goal for this project was to develop a program which promotes the inclusion of computation skills among students and help them work collaboratively with the use of human – robot interaction. Robots are such a strong tool that can be used to enhance computational thinking and engage students towards a technical field. Through workshops and readings about computational thinking we worked on building a block-based program that introduces the uses of robots as teaching tool for computational thinking.

Improving Semantic Document Classification Accuracy By Integrating Human-Crafted Knowledge, Zachary Weinfeld, Lubomir Stanchev

College of Engineering Summer Undergraduate Research Program

Document classification is a pivotal task in various domains, warranting the development of robust algorithms. Among these, the Bidirectional Encoder Representations from Transformers (BERT) algorithm, introduced by Google, has proven to perform well when fine-tuned for the task at hand. Leveraging transformer architecture, BERT demonstrates stellar language understanding capabilities. However, the integration of BERT with a range of techniques has shown potential for further enhancing classification accuracy. This work investigates several techniques that leverage semantic understanding to improve the performance of document classification models trained with BERT. Specifically, we explore three methods. First, we will balance corpuses afflicted by imbalanced …

The Methods Of Communication And Education Used To Train Forest And Fire Personnel Through The Swanton Pacific Ranch, Christine Brittney Johnson

Master of Science in Environmental Sciences and Management Projects

California does not have a cohesive curriculum for educating and training forest and fire professionals. The Swanton Pacific Ranch Fuels Management Training Program is Cal Fire affiliated and provides online and in person training to assist in equipping fire and forest professionals throughout California with the necessary skills to be successful fire and forest practitioners and managers. An exit survey was offered to all workshop participants after the Swanton workshop concluded. The survey was answered by 331 Swanton workshop participants to assess the effectiveness of the training program, the educational practices used to teach participants, and what factors of a …

Environmental Controls On Soil Organic Matter Fractions In Post-Fire Landscapes, Alexis Heuer

Master of Science in Environmental Sciences and Management Projects

In the past few decades, the world’s climate is being altered by anthropogenic forces such as the increase in carbon dioxide (CO2) emissions due to fossil fuel combustion (Paraschiv and Paraschiv, 2020). In California, the impacts of increased CO2 concentrations have created longer fire seasons, warmer temperatures, and increased lightning strikes which allow for wildfires to become more frequent, severe, and larger in scale (Westerling et al., 2006). Wildfires of this magnitude result in acceleration of landscape scale processes such as erosion and aggregation which in turn affects soil organic matter (SOM), the basis of soil health (Gonz´alez-P´erez et al., …

Fire Effects On Soil Organic Matter In The Creek Fire, Gracie E. Doolin

Master of Science in Environmental Sciences and Management Projects

Wildfires have increased in frequency and severity over the past few decades due to the increased concertation of CO2 emissions from anthropogenic influence. Soil carbon (C) sequestration has been identified as a climate change mitigation strategy; however, the influx of large-scale wildfires has accelerated landscape processes such as erosion, reducing soil aggradation, and soil C and nitrogen (N) protection. This trend is highlighted by the Creek Fire that occurred in September 2020 and burned 379,895 acres in the Sierra National Forest. This research is designed to close the knowledge gap regarding the impact of burn severity on soil organic matter …

Assessing Environmental Impact Of The Built Environment By Adopting Life Cycle Design: Using Swanton Pacific Ranch As A Case Study, Hope E. Springer

Master of Science in Environmental Sciences and Management Projects

To combat the carbon emissions contributed to the atmosphere by the built environment, it is imperative that low-embodied-carbon materials choices be prioritized throughout the building design process. To achieve this, this professional project creates an Excel-based, open source, and user-friendly tool for the construction sector to make design decisions that prioritize the sustainability of structures, as one such tool does not currently exist. The use of OpenLCA software and the EcoInvent34 database are utilized to calculate the environmental impact of 25 different building materials, as well as energy and electricity consumption. Swanton Pacific Ranch (SPR) in Santa Cruz County is …

Identifying The Stream Depletion Paradox By Monitoring A Stream’S Response To Aquifer Pumping From Neighboring Wells, James Klueber

Master of Science in Environmental Sciences and Management Projects

Current groundwater models utilize a constant head (Dirichlet) boundary condition which assumes stream stage is fixed and does not experience any drawdown in the event of pumping from an interconnected aquifer despite the presence of stream depletion. Therefore, constant head boundary implies that streams and lakes in a groundwater model behave as an infinite supply of water when aquifer pumping occurs. This study aimed to determine if a stream located in the California Central Coast experiences drawdown and depletion when pumping occurs within an aquifer-stream system. This was achieved by measuring stream stage, aquifer water levels, stream discharge, and the …

Greenhouse Gas Emissions And Fire Behavior Impacts Following California Fuel Treatments, Lindsay M. Senechal

Master of Science in Environmental Sciences and Management Projects

Fuel treatments are being increasingly applied across California landscapes as communities struggle to recover from wildfire disasters nationwide. Increased funding for treatments stems from grants under CARB’s Greenhouse Gas Reduction Fund and funding requires reports accounting the yielded benefits from these fuel treatments. Using data from real-world local scale fuel treatments, I used the forest simulation model FVSFEE and fire behavior software IFTDSS to quantify GHG emission benefits and fire behavior impacts 70 years after treatment. Results suggest that fuel treatments do not yield significant GHG benefits, and fire behavior impacts (conditional flame length) are minimal but overall show slight …

Establishing A Monitoring Framework To Evaluate Impacts Of Grazing Exclusion And Inform Restoration At Santa Rita Ranch, Annie Booth, Annie Booth

Master of Science in Environmental Sciences and Management Projects

Santa Rita Ranch is a 1,750-acre cattle ranch in Templeton, California, that was privately owned and continuously grazed for the past 70 years. In 2021, The Land Conservancy of San Luis Obispo (LCSLO) acquired the property, and in the winter of 2022, they built a fence that bisects the ranch, excluding the cattle from 800 acres to preserve the riparian habitat. The LCSLO is currently defining the ecological baseline for the ranch to inform future management decisions. Given that cattle grazing has many environmental tradeoffs, they wanted to establish a long-term monitoring study to understand how relieving the pressures of …

Pumping-Induced Drawdown Of Streams In Proximity To Agricultural Wells, Sara Alina Sternberg

Master of Science in Environmental Sciences and Management Projects

In the context of unprecedented drought and aquifer overdraft in California, understanding surface water and groundwater connectivity have become increasingly important. While there is a clear understanding of how wells induce drawdown in aquifers (Theis, 1935; Hantush, 1965), secondary effects on local streams have not fully been considered. Current mathematical models targeting aquifer drawdown do not address stream drawdown response in aquifer-stream systems with strong connectivity; instead, they assume the stream to be a constant head boundary. In the drought-stricken and summer-dry climate of California, streams often run dry for months out of the year, which is currently not reflected …

The Seoul National University Agn Monitoring Project. Iv. Hα Reverberation Mapping Of Six Agns And The Hα Size–Luminosity Relation, Hojin Cho, Jong-Hak Woo, Shu Wang, Donghoon Son, Jaejin Shin, Suvendu Rakshit, Aaron J. Barth, Vardha N. Bennert, Elena Gallo, Edmund Hodges-Kluck, Tommaso Treu, Hyun-Ji Bae, Wanjin Cho, Adi Foord, Jaehyuk Geum, Yashashree Jadhav, Yiseul Jeon, Kyle M. Kabasares, Daeun Kang, Wonseok Kang, Changseok Kim, Donghwa Kim, Minjin Kim, Taewoo Kim, Huynh Anh N. Lee, Matthew A. Malkan, Amit Kumar Mandal, Daeseong Park, Hyun-Il Sung, Vivian U, Peter R. Williams

Physics

The broad-line region (BLR) size–luminosity relation has paramount importance for estimating the mass of black holes in active galactic nuclei (AGNs). Traditionally, the size of the Hβ BLR is often estimated from the optical continuum luminosity at 5100 Å, while the size of the Hα BLR and its correlation with the luminosity is much less constrained. As a part of the Seoul National University AGN Monitoring Project, which provides 6 yr photometric and spectroscopic monitoring data, we present our measurements of the Hα lags of high-luminosity AGNs. Combined with the measurements for 42 AGNs from the literature, …

Functional Nanocomposite Coatings For Use In Food Packaging, Camden N. Webb

Master's Theses

Plastics are a class of materials known for their cost and property advantages, increasing significantly in their usage worldwide. Unfortunately, these benefits come with an increasingly concerning environmental impact. A combination of inadequate disposal options and combinations of materials have led to environmental disasters that will impact generations. One of the worst areas for plastic waste is food packaging. Plastic as a material generally excels at durability and longevity, but as food packaging, it outlives its intended purpose by several orders of magnitude. This leads to plastic food packaging materials sitting in landfill or leading to the environment for hundreds …

Covalent Adaptable Networks For Wood Coatings, Jachin Boaz Clarke

Materials Engineering

Wood swells and shrinks causing problems with seasonal humidity. Applying thick coatings of reactive finishes based on cross-linked polyurethane, epoxy, or polyesters can slow moisture-vapor exchange. However, the use of thick coatings leads to cracking and crazing sooner than thin finishes. This research proposes the addition of 3.3 mol % triazabicyclodecene, a conventionally used covalent adaptable network catalyst, in a commercially available polyester-based wood coating. The self-healing of the wood coating is tested using DMA stress relaxation and compression molding. The result from DMA renders inconclusive and compression molding indicates the novel wood coating oxidizes at elevated temperatures. The wood …

Detecting Coherent Transport Structures In Ocean Surface Flows, Leah Hoogstra

Master's Theses

Ocean surface transport plays a critical role in marine ecosystems, influencing the complex spatiotemporal patterns of both marine species and pollutants. The theory of Lagrangian coherent structures (LCSs) aims to identify fundamental patterns within time-dependent, nonlinear fluid flows. LCSs are material surfaces that act as dividing lines which fluid does not cross for a relevant period of time. LCS theory is still under active development, and there are multiple proposed ways to mathematically determine an LCS. Each proposed mathematical definition aims to capture the same physical properties, and some capture those properties more successfully and consistently than others. Here we …

A Novel Approach To Extending Music Using Latent Diffusion, Keon Roohparvar, Franz J. Kurfess

Master's Theses

Using deep learning to synthetically generate music is a research domain that has gained more attention from the public in the past few years. A subproblem of music generation is music extension, or the task of taking existing music and extending it. This work proposes the Continuer Pipeline, a novel technique that uses deep learning to take music and extend it in 5 second increments. It does this by treating the musical generation process as an image generation problem; we utilize latent diffusion models (LDMs) to generate spectrograms, which are image representations of music. The Continuer Pipeline is able to …

Analyzing Robustness Of An Agent Based Model On Action Potentials In Cardiac Tissue, Marion Jon Zollinger Lara

Master's Theses

An agent based model (ABM) is a computational model with ``agents'' that interact with each other in an ``environment.'' This paper analyzes a particular ABM simulating individual ions in cardiac tissue, with the goal of modelling the strength and consistency of the electrical signals needed for a healthy heartbeat. We build several frameworks based on work by M. A. Yereniuk and S. D. Olson to demonstrate robustness of the original model. We conclude a moderate level of robustness using those frameworks, through a combination of proofs and empirical evidence.

Exploring The Potential Of Pavegen’S Kinetic Energy Generating Floor For Sustainable Energy Solutions: A Proposal For Cal Poly Slo, Brandon J. Cuneo

Construction Management

This paper proposes the installation of Pavegen's kinetic energy generating floors at Cal Poly’s campus as a sustainable energy solution. Pavegen has developed a pioneering technology that converts footsteps into clean and renewable energy. The versatility of these floors is demonstrated through successful implementations in various settings, such as transportation hubs and public spaces, generating power from foot traffic. Collaborations with Schneider Electric, installation at Dupont Circle, and integration at Heathrow Airport showcase the potential for sustainable urban infrastructure. This paper outlines research conducted on Pavegen and similar solutions, including communication with company representatives and examining proposed installation locations at …

Neural Tabula Rasa: Foundations For Realistic Memories And Learning, Patrick R. Perrine

Master's Theses

Understanding how neural systems perform memorization and inductive learning tasks are of key interest in the field of computational neuroscience. Similarly, inductive learning tasks are the focus within the field of machine learning, which has seen rapid growth and innovation utilizing feedforward neural networks. However, there have also been concerns regarding the precipitous nature of such efforts, specifically in the area of deep learning. As a result, we revisit the foundation of the artificial neural network to better incorporate current knowledge of the brain from computational neuroscience. More specifically, a random graph was chosen to model a neural system. This …

Groups Of Non Positive Curvature And The Word Problem, Zoe Nepsa

Master's Theses

Given a group $\Gamma$ with presentation $\relgroup{\scr{\scr{A}}}{\scr{R}}$, a natural question, known as the word problem, is how does one decide whether or not two words in the free group, $F(\scr{\scr{A}})$, represent the same element in $\Gamma$. In this thesis, we study certain aspects of geometric group theory, especially ideas published by Gromov in the late 1980's. We show there exists a quasi-isometry between the group equipped with the word metric, and the space it acts on. Then, we develop the notion of a CAT(0) space and study groups which act properly and cocompactly by isometries on these spaces, such groups …

An Empirical Evaluation Of Neural Process Meta-Learners For Financial Forecasting, Kevin G. Patel

Master's Theses

Challenges of financial forecasting, such as a dearth of independent samples and non- stationary underlying process, limit the relevance of conventional machine learning towards financial forecasting. Meta-learning approaches alleviate some of these is- sues by allowing the model to generalize across unrelated or loosely related tasks with few observations per task. The neural process family achieves this by con- ditioning forecasts based on a supplied context set at test time. Despite promise, meta-learning approaches remain underutilized in finance. To our knowledge, ours is the first application of neural processes to realized volatility (RV) forecasting and financial forecasting in general.

We …

Analyzing Tortuosity In Patterns Formed By Colonies Of Embryonic Stem Cells Using Topological Data Analysis, Jackie Driscoll

Master's Theses

Pluripotent stem cells have been observed to segregate into Turing-like patterns during the early stages of Dox-inducible hiPSC differentiation. In this thesis, we de- velop a tool to quantify the tortuosity in the patterns formed by colonies of pluripo- tent stem cells using methods from topological data analysis. We use clustering techniques and the mapper algorithm to create simplicial complexes representing samples of cells and detail a method of evaluating the tortuosity of these complexes. We use the resulting persistence landscapes and their associated norms to evaluate experimental data and simulated data from an agent based model. This thesis finds …

Attainable And Sustainable Home Building Solutions, Katrina Matthews

Construction Management

Sustainability has fortunately been a recent concern in the construction industry as climate change continues to impose potentially insurmountable threats to the global environment. Although many new methods have been adopted in efforts to lessen the adverse environmental effects of construction-related activities, as a society, we still have much room for growth as the built environment accounts for almost 50% of global emissions. Many changes and standards must be adopted to make these solutions efficiently beneficial. To set industry standards, we must start with the most basic form of construction- residential building. This report covers two attainable framing solutions for …

Exploring The Mass To Light Ratio Of Massive Galaxies With The Rubin Dp0 Preview Dataset, Denvir Joy Higgins

Physics

The Vera C. Rubin Observatory will perform the 10-year long Legacy Survey of Space and Time expected to begin in 2024. LSST will cover the Southern Hemisphere, collecting over two million images with an 8.4-meter telescope and 3200-pixel camera. In collaboration with the scientific community and based on the Outer Rim Simulation \citep{heit19}, the Rubin Observatory released a simulated dataset (DP0) of the data that is expected to be in hand at the 5-year mark. Using this simulated dataset, I have explored the number and luminosity of the most massive local galaxies. Using a sample of the brightest, closest galaxies …

A Monte-Carlo Simulation Of Gamma Rays In A Sodium Iodide Detector, Ben Kessler

Physics

Gamma rays principally interact with matter through Compton scattering, photoelectric effect, pair production, and triplet production. The focus of this simulation is to study the theoretical energy spectrum created by gamma rays from a Cesium-137 source, which produces gamma photons with an energy of 0.662 MeV. At this energy level, most interactions are results of Compton scatters and the photoelectric effect. Therefore, this simulation only models those two effects on gamma rays. Using Monte Carlo methods and the Metropolis algorithm to sample the probability distributions of the two effects allowed for the simulation of gamma rays in a Sodium Iodide …

Diffusion-Driven Aggregation Of Particles In Quasi-2d Membranes, Oscar Gullickson Rausis

Physics

Many biological membranes can be modeled as two-dimensional (2D) viscous fluid sheets surrounded by three-dimensional (3D) fluids of different viscosity. Such membranes are dubbed quasi-2D as they exhibit properties of both 2D and 3D fluids. The Saffman length is a parameter that describes the energy exchange between the membrane and bulk fluids and controls the cross-over from 2D to 3D hydrodynamics. We aim to model diffusion-driven aggregation of particles embedded in a quasi-2D membrane. It is known that hydrodynamic interactions between solute particles significantly reduce their aggregation rate in 3D fluids. It is expected that in quasi-2D membranes the reduction …