24067 research outputs found
Sort by
GRM-076 Assessing the Performance of Intelligent Agents in Visual Food Recognition Relative to Manual Data Entry
Full text linkAccurate dietary assessment remains a critical yet time-consuming task in health and nutrition monitoring. This study benchmarks the macronutrient estimation capabilities of three intelligent vision agents: GPT Vision, Claude, and Gemini against manually logged food data. We unify two distinct datasets: MenuMatch, annotated by a professional nutritionist, and CGMacros, populated through user entries on MyFitnessPal. After flattening and cleaning both datasets, we first assess each model’s performance in calorie estimation. GPT Vision outperforms the others with the lowest percentage error 13.83% and is subsequently used to benchmark the macro estimations of Claude and Gemini. While Claude shows higher carbohydrate and fat estimation errors, Gemini yields the most balanced results across protein 12.55%, carbohydrates 19.57%, and fats 17.07%. These findings reveal strengths and trade-offs in current intelligent agents for visual food recognition, informing the development of more accurate, user-friendly, AI-powered nutrition tracking systems
GRM-080 Leveraging Data Science for Resilience: Improving Trauma-Informed Care Practice for Adverse Childhood Experience with AI & Data Science Application
Full text linkAdverse Childhood Experiences (ACEs) have long-lasting effects on physical health, mental well-being, education, and socioeconomic outcomes. Resilient Georgia (RG), a statewide initiative, seeks to address ACEs through trauma-informed care and data-driven strategies. However, challenges in data collection, analysis, and tracking hinder the effectiveness of these efforts. This study explores the role of data science and interactive visualization tools in improving outcomes for individuals and communities affected by ACEs. A key focus of this research is the development of a data science management application designed to enhance data collection and facilitate real-time decision-making. The application features interactive dashboards that allow stakeholders such as policymakers, healthcare providers, and community leaders to visualize program outcomes and track trends. Users can input data manually, upload files, and generate custom visualizations, making data more accessible and actionable for informed decision-making. By integrating these data-driven tools, RG can improve the monitoring and evaluation of its programs, optimize resource allocation, and ensure that vulnerable populations receive timely support. This study contributes to the growing body of literature on leveraging technology and data science to mitigate the impact of ACEs and promote long-term well-being
GRM-120 Coding Neurodivergent
Full text linkThis literary research provides a look at neurodivergent individuals learning coding, specifically Python, and the struggles and benefits that come from the way their brains are wired. This literature research was conducted to look at the benefits and struggles of learning to code as a neurodivergent individual. One area that has not been studied extensively is the learning of Python, or coding in general, by neurodivergent populations. This includes the benefits a neurodivergent learner may glean from learning Python as well as the challenges they may encounter - anything which makes their experience different from that of a neurotypical Python beginner. Outcomes for which data was sought included cognitive challenges neurodivergent individuals may face (such as difficulty understanding abstract concepts or issues with motivation), ways in which the neurodivergent brain may be well-suited to learning Python, and ways in which neurodivergent populations may benefit from learning Python
GRM-131 XR Agent (A MLLM powered XR system)
Full text linkThis project proposes “XR Agent”, a uncoupled and efficient framework for developing AI-powered extended reality (XR) applications on head-mounted displays (HMDs). Leveraging multimodal artificial intelligence—including MediaPipe(Google open-source CV Model) for computer vision (object segmentation, recognition, pose estimation), multimodal large language models (MLLMs) like Gemini, and Unity’s cross-platform XR development ecosystem—the framework aims to create an extensible base system that enables rapid prototyping and deployment of intelligent XR applications. Currently, it was deployed on the Meta Quest 3 platform, XR Agent explores novel HCI(Human Computer Interaction) paradigms, combining real-time sensor data processing, immersive visualization, and adaptive AI-driven logic. This work addresses challenges modular integration of various different kinds of devices AI models. The framework also will be valuable through use cases in collaborative remote control, immersive training scenarios, and data collection for embodied AI
UC-027 KSUBlocks Tower Defense
Full text linkOur project, KSUBlocks Tower Defense, is a Minecraft Plugin designed to create a game mode in the Tower Defense genre. We aim to create a unique and fun game for the students in the KSU Minecraft server. Our project is entirely configurable allowing for easy maintenance and room for future expansions, while ensuring the server performance remains steady alongside KSU\u27s other game modes. It is developed in Java, utilizing IntelliJ and Paper API. We plan to deploy it on the KSU Minecraft Server upon finalization
UR-017 SHERLOCK: Self-supervised Histopathological Evaluation for Recognition of Lymphocytes and Other Cancerous Kinds
Full text linkWhole Slide Images (WSI) are gigantic images (e.g. 100k x 100k pixels) of tissue samples. The goal of SHERLOCK is to detect cancer cells in those tissue samples. We do this by using a pretrained Masked Autoencoder (MAE), from Facebook’s research lab, that we finetune on the PanNuke dataset. The benefit of using an MAE is that unlike supervised learning the WSI’s don’t need to be labeled. This is important because it will save a lot of time and money that would be spent on labeling WSI’s
Machine Learning Approaches in Sustainable Maintenance Practices
No full textThe cost of maintenance activities generally accounts for the second largest portion of the operational budget after energy costs in an industrial ecosystem. The significant reason behind the continuing rise of maintenance costs reflects on the application of highly automated and technologically complex machineries involved in the manufacturing process. As such, the implementation of sustainable maintenance practices, in the realm of Industry 5.0 and the integration of Machine Learning (ML)/Artificial intelligence (AI) methodologies, is essential more than ever for enterprises to predict and avoid system failures and to subsequently save cost and sustain in the business. In this work, we investigated the ML/AI approaches and the importance of holistic human-machine integration in industrial facilities. Additionally, we will create a framework for human-centric methods of integrating automation into sustainable maintenance practices where human intervention remains crucial in interpreting complex data, making judgment calls in uncertain scenarios, and applying context-specific knowledge that machines alone cannot provide. By placing humans at the center of our framework, businesses can ensure a more holistic approach to maintenance that accounts for both technological advancements and the irreplaceable value of human ingenuity while still taking full advantage of the remarkable potential of ML methodologies and AI-driven technologies as a viable process for achieving reliability and productivity. In the next step, we intend to offer such a designed framework to local manufacturing facilities in Georgia, and to analyze the responses from a pilot test conducted in that environment
Crafting Appeal: An Analysis of Harper & Brothers\u27 19th-Century Publishers\u27 Bindings
No full textThe United States is well known as a dominant force in the global publishing industry. The long trajectory of this success began during the early nineteenth century when the production and publication of printed books entered a new era. Prior to the nineteenth century, all elements of printed books were crafted by hand and were expensive to acquire. During the late eighteenth and early nineteenth centuries, the domino effect of industrial advances revolutionized and economized book printing and publishing, providing common people access not only to books, but attractive ones. At the forefront of this advancement was the Harper & Brothers Establishment, today known as Harper or HarperCollins Publishers. Harper & Brothers was one of the first publishers to incorporate an in-house bindery, jet setting them miles ahead of their competitors. This bindery produced their famous monthly publication, Harper’s Monthly, and their weekly publication, Harper’s Weekly or Harper’s Bazaar, which provided invaluable marketing for their published books and novels. But what put Harper & Brothers on top was their use of handsome book bindings, particularly the use of “library sets” during the Golden Age of Book Collecting to attract and retain a consistent market of book buyers during the nineteenth century. This study analyzes a collection of the Bentley Rare Book Museum’s Harper Establishment bindings from 1833 to 1900 to understand how Harper used decorative and highly curated sets of bindings to attract book buyers during a competitive moment in book history. This study is significant because it demonstrates the historical importance of publishers’ bindings during the nineteenth century and sheds light on twenty-first century book production practices that continue these traditions
Division in Resistance: Georgia and Massachusetts After the Intolerable Acts
No full textThis project discusses the colonists’ reaction to the Intolerable Acts and how the differences across New England and Southern colonies reflect their colonial identity. Previous scholarship discusses the resistance to the Intolerable Acts through perspective focused on economics, grassroots, or ideological discourse. I plan to break away from this by focusing solely on how colonial identity shaped their resistance strategies. Throughout New England, there was a long history of political activism, so they were quick to respond and organize resistance movements. In the Southern colonies however, there was a reliance on Britain within their economy, so they were much more hesitant to organize resistance movements. By analyzing arguments from authors such as Bernard Bailyn, T.H. Breen, Pauline Maier, and others, this project places the records of the colonial reactions to the Intolerable Acts in a broader perspective of what we can tell about the regional political differences. This project also contributes to the debate on how popular the resistance movements were, as though the resistance movements were widespread throughout the colonies, they were not a cohesive unit. These regional differences would come to play a significant role in the development of the Republic in the early years as well, and are also reflected in modern politics. By using primary sources, such as colonial newspapers, town meeting records, and petitions, this paper breaks down the regional differences between New England and the South during the late 18th century
What\u27s Your Story? Narratives and Career Development
No full textPeople have many different reasons why they choose to pursue certain careers. In our research, we are gauging how the use of narratives influences our decisions on what career paths we choose to take. Based on the research we conducted for the purpose of this study, we have found that the use of narratives often does have a direct impact on how people choose to approach their career paths. Sometimes these impacts push them toward the career path they’ve been told about while other times it steers them away and in other directions. To get a bigger picture of what exactly these impacts are, we asked a few critical questions: What are some stories that you’ve heard in your life that have shaped your career direction? What are your major and/or career goals? Is there someone who has inspired you to make this decision, and how so? These questions have brought about a great deal of conversation from different perspectives, some whose stories led them down a certain path and others who have only been pushed in a direction, but not onto a set road. We conducted our research in a methodical way by interviewing students and holding focus groups, taking into account gender, racial, ethnic, geographic, class, and age-related differences. In these interviews and focus groups, we have found that the environment in which people grew up, the interests or career paths of those around them, as well as the stories that they’ve heard while growing up all have a major impact on their future career path. We are continuously attempting to understand how narratives impact the career choices of people and will continue to interview people to get a better view of it. We know that stories have a powerful impact on career choices and that the way that they are communicated is equally as powerful. These results will be presented at the Spring Symposium of Student Scholars