ScholarWorks (California State University)
Not a member yet
87523 research outputs found
Sort by
FacTRAK: Randomized Kaczmarz Algorithm for Tensor Systems with Factorization and Averaging
Get PDFRandomized iterative methods, such as the randomized Kaczmarz algorithm, have become increasingly popular for their efficiency in solving regression problems involving matrices. However extending their application to tensor regression - despite its natural occurrence in many fields - is still in it's infancy. As data becomes increasingly complex, tensors emerge as powerful data structures for tasks such as multi-dimensional data analysis. In this work, we introduce the Factorized Tensor Randomized Averaging Kaczmarz (FacTRAK) algorithm, an extension of the iterative method tailored for solving randomized averaging tensor linear systems within the Kaczmarz algorithm using the t-product framework. We decompose the traditional tensor randomized averaging Kaczmarz method (TRAK) to extend over subsystems of a consistent tensor system. Leveraging insights from randomized Kaczmarz techniques, we develop and analyze this method to address the unique computational challenges posed by tensors
Enhancing Zigbee Communication Efficiency using FELACS Data Compression in Python
No full textThis thesis proposes the design and performance test results of FELACS, specially tailored for efficient data over wireless transmission in Zigbee-based WSNs in the context of IoT. The increasing deployment of IoT applications has raised critical needs for an efficient data-handling approach in a resource-constrained environment. FELACS uses a neighborhood indexing sequence algorithm that captures the spatial and temporal correlations among the sensor data, significantly improving compression ratios, reducing the transmission time, and thus consuming low power. Experimental results show how the algorithm can achieve compression ratios up to 75% for small data packets, which effectively addresses bandwidth limitation challenges and energy efficiency in real-time applications. Results of the current work highlight that FELACS represents a valid and effective solution to improve Zigbee radio communication in various IoT application scenarios. Ensuring a flexible and adaptive compression framework, FELACS provides improved transmission efficiency, along with assurance of data integrity in susceptible applications related to environmental monitoring and health care. This research paper adds to the literature by adding a new dimension to the field of data compression in Zigbee networks. It provides some future directions on how integrating hybrid compression techniques and enhanced security will provide more robust and efficient IoT systems
Flipping the Deck: Transforming Ace Industrial Supply's Public Perception
Get PDFOur research indicated that Ace's sales model must be updated to remain relevant and compliant with current business practices. Ace's current sales model involves significant amounts of overhead, and Ace's policies do not promote their employees engaging in moral and ethical sales practices. To solve this issue, we have developed a system of KPIs and goals, along with an incentivization structure, which allows salespeople to maintain their current salaries if they are engaging in the ethical and sustainable sales practices. The Ace website must also be redesigned to be more use-friendly and allow other avenues of sales. Without these changes, Ace will have no way to repair their public image and continue to grow into the future. As a long-term plan, we have recommended that Ace build on this growth by creating a consolidated call list of leads, loyalty benefits for returning customers, text/email order verification and AI support in finding quality leads. For this project, I provided analysis of Ace's financial data and internal survey results, including organizing and formatting the data into a form that is understandable by the team. I completed a PESTEL analysis of Ace as well as the external threats and part of the internal weaknesses for the company. I spent significant time data mining and analyzing internal information using Ace's internal data analysis tool, and assisted my teammates in proofreading and improving the quality of writing of their sections, as well as brainstorming long term solutions for improving Ace and helping to create the overall layout of our paper. I spent between 6 and 10 hours a week on this course, similar to the remainder of my team, for weeks 1-5 of this course and spent 15-20 hours a week on it since
Forbidden Nature
No full text[ABSTRACT ONLY; NO FULL TEXT] Alexa has lived a simple life of magic since the day she was born. She is the most important witch in existence, the Chosen One. Then one day everything suddenly changed. The life of a not so simple witch was about to get a little more complicated. A team of vampires went out with a single objective in mind, to kill Alexa DeMarco. But instead unknowingly turned her. Ever since Alexa has been trying to curve the bloodlust with the help of her best friend and protector, Zach. He's a werewolf born into a prestigious circle of werewolves, called the Rulcaunts, where their only goal in life is to protect the person they are appointed to. But Zach has alot to prove towards Alexa being his top priority, he should have been there the day of the attack, he was missing for some time before he showed his face or fangs to protect her
Walking With Evil: Puritan Sentiments in Hawthorne's Reimagined Wilderness
No full text[ABSTRACT ONLY; NO FULL TEXT] Attitudes concerning wilderness have historically been influenced by cultural, religious, and ideological perspectives, each shaped by the time and place in which they exist. This essay seeks to redefine the concept of wilderness by examining biblical allusions in Nathaniel Hawthorne's short fiction "Young Goodman Brown," while also contextualizing the term within the framework of both Hawthorne's Puritan ancestry and the unique historical period in which he lived. Scholars such as Reginald Cook have often interpreted Hawthorne's writing through a jeremiad lens, depicting his wilderness as a "primitive" entity that exposes the inherent evil and sinful nature of humanity. Conversely, Leo Levy argues that Hawthorne portrays a "sublime acceptance" of the landscape, therefore empowering him to use wilderness as a tool that challenges Puritan notions of evil and sin, while also passionately embracing these moral complexities. Michael Bell offers a more "balanced" perspective, characterizing Hawthorne's wilderness as a space where both distrust and transformation coexist, acting as a divide between Puritan sentiments and the problem of inherent evil, yet offering readers an opportunity for introspection. I define Hawthorne's wilderness as a self-contained, autonomous entity separate from the acts occurring in the forest. Instead, wilderness serves to instruct the colonizer mindset, therefore advocating for its own natural value and allowing Goodman Brown the opportunity to confront and discover his own moral conflicts. Wilderness, then, is not an "evil" needing to be tamed or saved, rather wilderness embodies the qualities of a teacher, revealing the moral incongruencies embedded within human nature itself. Considering Hawthorne's work through an ecocritical lens positions audiences to reimagine his dark wilderness as a redemptive landscape, therefore allowing readers to experience a type of moral sublimity that reflects Hawthorne's exploration and divergence of the Puritan concept of evil and the ethos of American Romanticism. By illuminating Hawthorne's close kinship with his environment and juxtaposing it with the Puritan notions of wilderness, modern readers will develop a deeper and more nuanced appreciation not only of Hawthorne's own attitudes toward environment, but also how these viewpoints resonate broadly as we continue to contemplate, define, and reshape our modern understanding of wilderness
Well Worth the Price of Admission: San Diego and the Origins of the First Integrated Professional Baseball League in the United States, 1910-1916
Get PDFThe story of integration in professional baseball is traditionally focused around the debut of Jackie Robinson with the Brooklyn Dodgers in 1947. The emphasis on Robinson, deserved as it may be, has obscured the long and complicated history of the Black players in baseball who preceded him. In recent years there have been concerted efforts by historians and researchers to shine more light on this era of baseball history. This thesis fits neatly within these modern efforts. The thesis takes a close look at the origins of integration in the California Winter League, which was the first continuously integrated professional baseball league in the United States. It uses contemporary newspaper accounts from 1910 to 1916 to document the impact this period had on the different parties involved, from the city of San Diego to the Chicago American Giants, the Black baseball club asked to participate
Achieving High Factor VII Knockdown Efficiency In Vivo with Manual Mixing siRNA-Lipid Nanoparticle (LNP) Formulation
Get PDFThe goal of this experiment was to compare manual mixing method to microfluidic mixing with the delivery of factor seven (FVII) small-interfering RNA (siRNA) using a lipid nanoparticle (LNP) liver reagent offered by Thermo Fisher Scientific (TFS). The Clinical Delivery Team at Thermo Fisher Scientific, Carlsbad aims to achieve high FVII protein knockdown with manual mixing to offer customers, without a microfluidic device, an efficient and effective solution to delivering siRNA or mRNA. Currently, microfluidic mixing is the industry gold standard in the LNP field as this method is robust and formulates consistent and stable LNPs. However, manual mixing of siRNA-LNP formulation allows many customers that don't have access to expensive microfluidic instruments the opportunity to efficiently and effectively screen experiments with Thermo Fisher Scientific LNP liver reagent. The research performed, in vivo with Balb/C mice, compared 1. (physio characterization of the LNP formulations), 2. (FVII knockdown in plasma), 3. (liver enzyme levels), 4. (immunogenicity response), and 5. (body weight tracking). Results indicate manual mixing of siRNA and TFS LNP liver reagent may just be as effective as microfluidic mixing. These findings suggest that manual mixing of LNPs could potentially serve as an alternative and viable approach for customers to deliver siRNA in vivo. The Clinical Delivery Team at Thermo Fisher Scientific, Carlsbad needs to further explore and understand the LNP liver reagent to provide customers with the best LNP product and experience in the market
Tailored Environmental Monitoring for a Cell Therapy R&D Lab
No full textObjective: Thermo Fisher Scientific, Inc.'s cell biology lab 5055 is a shared R&D, biosafety level (BSL) 2 lab. The lab includes 20 shared aseptic biosafety cabinets (BSCs), 6 shared incubators, and is used by dozens of scientists in a 1500 sq. ft. space. Projects in lab 5055 are sensitive to contamination, which can impact experiments, budgets, and timelines. As a BSL 2 lab, it has not been held to microbial growth limit standards. This project aims to design an environmental monitoring (EM) program using FDA and ISO guidelines tailored to lab 5055's unique space and function, integrating seamlessly into existing maintenance workflows. Methods: Using FDA and ISO guidelines, layout maps of the lab were created and overlaid with heat maps showing key parameters like foot traffic and critical areas. ISO classifications were assigned to spaces such as BSCs and supporting areas for future microbial monitoring. An Excel-based sampling plan generator was developed to randomly select 50% of the lab space for monitoring. The lab was monitored for 3 months using passive air and surface contact sampling with Tryptic Soy Agar (TSA) plates. Colony forming units (CFU) were enumerated and compared to regulatory recommended standards. Results were visualized on layout maps, identifying problem areas. Recommendations for protocol changes were made to reduce microbial growth. An online toolset containing the sampling generator, data, protocols, and supplementary literature was published on the Thermo Fisher intranet. Results: The online tools help provide auto-selection of sampling and formatted data organization, ensuring data traceability and historical tracking. The sampling methods are cost-effective and easy to use, allowing any scientist to perform the EM. The condition of the lab is visualized in date-specific layout maps, facilitating communication about the current work environment. Recommended changes to current lab cleaning and usage protocols aim to help reduce and manage microbial growth. Conclusion: The EM program is specific to lab 5055's space and function, fitting well into current maintenance workflows. Three months of monitoring helped to provide valuable microbial growth data, enhancing contamination communication. This project lays the groundwork for a more through EM program for lab 5055. The developed data management, online toolset, and protocols allow for future improvements, increasing the program's value to lab 5055 scientists and the company
Estimating the Health Quality of Campus Meals Using Artificial Intelligence
Get PDFCampus dining plays a significant role in influencing student health, academic performance, and overall well-being. Many campus meal options do not provide easily accessible nutritional information at the point of purchase, which makes it challenging for students to make informed food choices. This study investigates the design and evaluation of three artificial intelligence (AI) approaches aimed at classifying and offering nutritional insights for campus meals. These approaches include: (1) Object detection using Roboflow/YOLOv8, (2) Image classification using MobileNetV2, and (3) optical character recognition (OCR)-based menu text analysis using Tesseract. A custom-built dataset was created using publicly available nutritional data from Nutritionix and images sourced from campus restaurants and academic collaborators. Each AI method was evaluated based on its ability to accurately identify food items and classify them as healthy, moderately healthy, or unhealthy based on a holistic set of nutritional metrics. The object detection model achieved the highest mean Average Precision (mAP@50) of 79%, while the OCR-based method reached a dish-level detection Accuracy of 85.94%. This research demonstrates the feasibility of AI-powered systems to facilitate real-time nutritional assessments and meal classifications. It contributes to the development of accessible health technology for university students, promoting better dietary choices and improving health outcomes
The Monsters of Mourning
Get PDFThe Monsters of Mourning is a multimodal and multi-genre creative thesis, using memoir and poetry, along with different art forms, to express my own lived experiences with grief and loss by using the voices of cryptids and monsters from international folklore and mythology that are connected to our worldly shared fear of losing people