University of Central Florida
University of Central Florida (UCF): STARS (Showcase of Text, Archives, Research & Scholarship)Not a member yet
166656 research outputs found
Sort by
Transforming Health Science Education: AI-Generated Video Case Studies and Clinical Makeup Assignments
No full textHealth science students in didactic and clinical courses excel with application-based learning that enhances clinical judgment and critical thinking. AI-generated case studies and videos bring dynamic, immersive simulations into the classroom, bridging the gap between theory and practice while boosting engagement and preparedness
Multidimensional Approaches in Bug Detection for Parallel Programming and Text-to-Code Semantic Parsing
Get PDFThis dissertation applies deep learning and large language models to two domains: parallel programming fault detection and text-to-code translation, aiming to enhance software reliability and natural language-driven code generation. Due to their unpredictable nature, concurrency bugs-particularly data race bugs— present significant challenges in fault detection for parallel programming. We investigate deep learning and LLM-based approaches for detecting data race bugs in OpenMP programs. Our proposed methods include a transformer encoder and GPT-4 through prompt engineering and fine-tuning. Experimental results demonstrate that the transformer encoder achieves competitive accuracy compared to LLMs, highlighting its effectiveness in understanding complex OpenMP directives. Expanding this research, we explore the role of LLMs in detecting faults in Pthreads, which requires a deep understanding of thread-based logic and synchronization mechanisms. We analyze ChatGPT\u27s effectiveness in Pthreads fault detection through dialogue-based interactions and advanced prompt engineering techniques, including Zero-Shot, Few-Shot, Chain-of-Thought, and Retrieval-Augmented Generation. Additionally, we introduce three hybrid prompting techniques—Chain-of-Thought with Few-Shot Prompting, Retrieval-Augmented Generation with Few-Shot Prompting, and Prompt Chaining with Few-Shot Prompting—to enhance fault detection performance. In the semantic parsing domain, our research bridges the gap between natural language and executable code, focusing on text-to-SQL translation. To address SQL\u27s limitations in statistical analysis, we introduce SIGMA, a dataset for text-to-code semantic parsing with statistical analysis capabilities. In addition, we address the gap in cross-domain context-dependent text-to-SQL translation for the Arabic language. While prior research has focused on English and Chinese datasets, no efforts have been made to explore Arabic cross-domain conversational querying. We introduce Ar-SParC, the first Arabic cross-domain, context-dependent text-to-SQL dataset. This dissertation contributes to fault detection in parallel programming and semantic parsing with statistical analysis, leveraging cutting-edge deep learning and LLMs techniques. Our findings advance bug detection in high-performance computing and natural language-based code generation, significantly improving software reliability and accessibility
Development of a Laser Spectroscopic Device for Measurements of Temperature and Speciation in Detonation Environments
No full textCharacterizing novel detonation-based combustion systems is critical to assessing new designs and improving the understanding of their unique operations. Due to the harsh nature of a detonation environment, industry-standard sensing and diagnostic tools can fail to provide adequate methods for in-situ quantitative measurements of key thermodynamic properties. Diagnostic equipment must adapt to meet the needs of new combustion systems. Laser absorption spectroscopy (LAS) is a technique commonly utilized in combustion research to provide such measurements in laboratory settings due to its precision, minimally invasive properties, and high data acquisition rates. To prepare such a sensor for use in a challenging environment like a detonation reaction, considerations must be made to adequately address potential sources of error or sensor damage such as harsh operating conditions, intense vibrations, absorption non-idealities, and interfacing concerns that are often not experienced in a laboratory setting. A LAS sensor can easily be incorporated into facilities where state-of-the-art detonation combustors are tested. A LAS sensor was developed, utilizing fixed wavelength tunable diode laser absorption spectroscopy to perform measurements of three target species within a state-of-the-art rotating detonation combustor (RDC). A second iteration of the sensor was also developed with numerous improvements to both measurement capabilities and sensor hardware, which was developed utilizing the Scanned LAS technique to perform a second set of measurements. This document covers the development of the sensor, its application within a state-of-the-art RDC, and the results of the diagnostics performed
Smart Work Zone Connectivity for Traffic Safety Applications
No full textWork zones play a vital role in the maintenance and upgrading of roads and highways; however, they also pose substantial risks to both drivers and workers. Utilizing technology, such as Smart Work Zones (SWZs), Connected Autonomous Vehicles (CAVs), and data dissemination algorithms, is crucial in improving safety in work zones by providing decision-makers with timely and accurate information. Ensuring safety in these environments is essential for reducing accidents and fatalities and enhancing drivers\u27 sense of security, and protecting the well-being of workers operating in hazardous conditions. In this dissertation, we proposed a methodology based on three complementary approaches to enhance work zone safety. First, the study develops a cost-effective, low-power communication network for SWZ barrel-mounted sensors, proposing a novel relay node selection algorithm utilizing Bluetooth Low Energy mesh technology to enhance network reliability in linear, semi-static configurations. Second, the impact of vehicle-to-vehicle communication on lane change safety is assessed through a calibrated co-simulation framework, modeling real-world connectivity issues such as packet loss and latency to analyze their effect on CAV behavior during merging maneuvers. Third, a Signal Temporal Logic-based optimization framework is introduced to dynamically enforce safety constraints and refine lane change parameters, addressing the limitations of traditional car-following models in complex traffic environments. Finally, the research integrates vehicle-to-infrastructure communication into a comprehensive SWZ framework, evaluating the combined effects of CAV market penetration rates, roadside unit placement, and communication ranges on crash risk reduction. Together, these contributions create an interconnected framework for improving traffic safety in dynamic and complex work zone environments. The integration of these methodologies not only bridges gaps between traffic behavior modeling and communication network analysis but also provides scalable, real-world solutions for enhancing safety and efficiency in smart transportation systems
Small Ripple, Big Waves: Individual Political Attitudes, Populism, Illiberalism And Democratic Resilience
No full textWhy do individuals who support democracy also endorse populist and illiberal attitudes? Current literature focuses efforts on exploring the role that institutions and political actors have in democracy resilience (Levitsky & Ziblatt, 2019; Merkel & Lührmann, 2021). However, relatively few academic works have examined this phenomenon at the individual level. With the aim to exploring the causal effect between individual political attitudes, anti-democratic threats, and democratic resilience at the individual level, this study applies a micro-level theoretical approach to assess how individuals’ political attitudes toward populism and illiberalism affect their support for democratic values. To test the hypotheses, this study analyzes data from the seventh wave (2017-2022) of the World Values Survey cross-analyzing the cases of Brazil and the United States. The main findings indicate that institutions and political actors have a substantial impact on individual political behavior, which, in turn, shapes support for democratic values. Ideological alignment also influences tolerance for anti-democratic actions, with political polarization exacerbating threats to democratic norms, especially in the United States. While this study is, to some extent, limited by endogeneity issues and survey constraints in capturing the full complexity of political behavior, it contributes to the literature by emphasizing the role of individuals in democratic resilience. Future research should further explore how political engagement and societal mobilization interact to either sustain or erode democracy
Increasing Bystander and Environmental Awareness Through Haptic Feedback
No full textThe Chariot System is a bystander and environmental awareness system that uses vibrotactile haptic feedback delivered via a wearable. It was designed with the goal of using a subject\u27s haptic channel to keep them aware of their surroundings thus freeing their visual channel to focus on and enjoy a virtual reality (VR) experience. This work contributes to the space of VR research that aims to keep users aware of the outside world while in VR, an area that typically focuses on using visual notifications but rarely applies haptics. A user study involving 24 participants was conducted comparing the Chariot System to an analogous visual notification system to determine haptic feedback’s efficacy of alerting participants to virtual passersby while partaking in two different but common VR tasks, watching a video and playing a VR game. After going through each condition, participants filled out the NASA-TLX and System Usability Scale surveys to gauge their experience with both systems. The survey results as well as their accuracy and response times to notifications were then analyzed to compare the performance of both systems. Our findings show that while reaction time using the visual notifications was faster, participants found the use of the Chariots less mentally taxing and preferred. This work could possibly be expanded upon in the future through the addition of real time sensors and more portable power sources for the wearables such that the VR user could physically move around in the virtual environment and be made aware of them approaching or being approached by oncoming persons or obstructions
Screen Time on Stage: The Effect of Internet and the Smartphone on the Actor
No full textThis thesis examines the potential link between an actor’s cell phone use and their work in the rehearsal process. Using Paula Thomson and S. Victoria Jaque’s Creativity and the Performing Artist and Nichola Carr’s The Shallows: What the Internet Is Doing To Our Brains, I analyze the neural networks of the brain that are engaged in acting and their functions to develop a greater understanding of their association to behaviors and emotional responses that may impact the actor in the rehearsal/performance space. Furthermore, I investigate the emergence of smartphone use within the cultural zeitgeist and its effect on the actor’s behaviors to postulate how this technology may influence the actor at work. In order to examine smartphone use as it relates to my own experience, I keep a journal over the course of a three-week period during Pegasus Playlab and a six-week rehearsal/performance period of Macbeth at Orlando Shakes. I document moments of impact, my ability to connect with actors on stage, my ability to focus, and my overall mood throughout the day. During my time with Playlab, I turn off my phone during the beginning of the rehearsal period and don’t turn it back on until the day’s responsibilities are over. I will run the same experiment during my time at Orlando Shakes with Macbeth, this time running a bi-weekly schedule of “Phone on/Phone off” work periods. At the end of six weeks, I will reflect on what differences were present (if any) between weeks with and without a smartphone in the room, how they were similar or different to my previous experience at Playlab, and whether a change in smartphone behavior is worth incorporating into my process
Flywheel Technology: The Reliability of Functional and Systemic Perturbations in a Crossover Design
No full textThe purpose of this dissertation series was to evaluate the reliability, validity, and applicability of the Desmotec D.EVO isoinertial flywheel device (D.EVO) for both isometric and dynamic squat assessments. Study one sought to establish the validity of the D.EVO for assessing squat maximal voluntary isometric contraction (FMVIC) versus a traditional squat maximal voluntary isometric contraction (TMVIC), determine the test-retest reliability of force and power outputs during FMVIC and flywheel squat protocols (FSPs), and explore relationships between these outputs and broad jump (BJ) and vertical jump (VJ) performance. We also identified the number of familiarization sessions needed for reliable data collection. Results showed FMVIC was reliable after two familiarizations, while FSP was reliable after four familiarization sessions. Despite strong associations between FMVIC and TMVIC, agreement between the two modalities was poor with FMVIC significantly overestimating force. FSP concentric average peak power was the only predictor of BJ and VJ performance. Study two examined differences in participant pain and force production during full-body harness (FBH) and waist belt (WB) FMVICs on the D.EVO, assessing the reliability and validity of these protocols compared to TMVIC. The results of the study indicated that although the FBH was more tolerable to wear during a squat MVIC, both straps significantly overestimated force and, therefore, neither exhibited acceptable agreement with TMVIC. Study three characterized the magnitude, timescale, and reliability of functional and systemic markers of exercise induced muscle damage (EIMD) following moderate and high isoinertial resistance FSPs, provided commentary on the presence of the repeated bout effect (RBE) in resistance-trained men, and offered recommendations for utilization of the FSP in crossover studies. The results of the study indicated a negligible presence of the RBE in this population, indicating that the FSP may be utilized in cross-over designs provided the study sample is of sufficient resistance training status
Laminar Burning Speed of DME-hydrocarbon Mixtures
No full textFossil fuels, when burned in the air, release large amounts of carbon dioxide (CO2) and greenhouse gases (GHG) into the air, causing significant pollution. To reduce greenhouse gas emissions (CO₂, CO, and NOₓ), alternative fuels such as Dimethyl Ether (DME) are being explored for their potential to offer cleaner combustion solutions. This study investigates the laminar burning speed (LBS) DME blends with several alkanes and alkenes in air at atmospheric pressure (1 atm) across a wide range of equivalence ratios (Φ). LBS is essential for understanding combustion characteristics, including combustion efficiency, heat release rates, and chemical kinetics. Accurate LBS measurements are vital in optimizing combustion system designs, such as those used in gas turbines and internal combustion engines, where efficient fuel burn and minimal pollutant formation are critical. To comprehend the performance and usability of these fuels in gas turbines, measurements of parameters like ignition delay time and laminar burning velocity (LBV) are required. DME-propane mixtures are exciting in this study due to their potential for enhanced combustion stability and flame propagation. Here, we report data on observations of these fuel’s laminar burning speed up to an initial pressure of 1 atm and an initial temperature of 295 K. The LBS measurements were performed using a spherical chamber with Schlieren imaging techniques to visualize the flame front and ensure precise tracking of flame propagation. The unburned gas was maintained at ambient temperature, with precautions to minimize heat loss. Results indicate that LBS varies significantly with the equivalence ratio, reaching its maximum near stoichiometric conditions for all fuels. As part of the validation procedure for the results, they are also contrasted with the performance of comprehensive kinetic models
Of Fate and Fear: An Interactive Narrative Approach to Experiential Horror
No full textThis thesis examines the nature of guest interactivity in themed entertainment, considering best practices when approaching the creation of an interactive theme park offering with the belief that narrative interactivity and guest agency increases the potential for meaningful, memorable, and engaging entertainment experiences. Taking both established literature and existing offerings into consideration, this thesis presents an original creative narrative design concept for a Halloween haunt event overlay, titled Fear the Unknown, that enhances the guest experience with an emphasis on interactivity and agency. This creative work is presented specifically from the point of view of the show writing discipline, consisting of a high-concept treatment and other relevant narrative design documentation. The setting of a Halloween event was chosen for its uniquely immersive, narrative, and visceral nature. For this exploration, it is viewed as a relatively self-contained offering where an interactive narrative could increase guest engagement in the event’s story and overall experience, thus driving greater and repeat attendance without sacrificing the practicalities and efficiency demanded of a major theme park event. Logline: When a sinister cult of hackers unleashes their techno-terror on a local theme park haunt, you must choose your fate and fight for survival in this interactive, story-based horror event. Do you Fear the Unknown