University of South Alabama Institutional Repository
University of South Alabama Institutional RepositoryNot a member yet
5383 research outputs found
Sort by
Does AI Intergration Moderate the Relationship Between Firm Growth and Performance in SMES: The Influence of Decision-Making and Operational Performance
Full text linkToday\u27s dynamic business environment requires small and medium-sized enterprises (SMEs) to keep up with technological advancements in order to remain competitive. Business growth creates more challenges for SMEs since they possess fewer available resources than big organizations. Since the introduction of artificial intelligence (Al), several SMEs have been able to compete more effectively and deliver better performance. As part of this research, I examine the possibility that Al integration (All) will moderate the relationship between firm growth and both decision-making and operational performance, ultimately affecting the performance of SMEs. The aim of this research is to provide practical implications for Al as a strategic resource for improving decision-making capabilities, performance and growth by utilizing the resource-based view (RBV) and information processing theory (IPT). A partial least squares structural equation model (PLS-SEM) was used to analyze data from 338 SME business strategy decision-makers in the United States. In order to verify the measurement model\u27s reliability and validity, a Confirmatory Composite Analysis (CCA) was performed, followed by the evaluation of the structural model in order to test the hypotheses. In contrast to initial hypotheses, this study found that firm growth is positively related to both decision-making and operational perfonnance. Nevertheless, the study results support the original hypothesis that both decision-making performance (DMP) and operational perfonnance (OPP) positively affect a finn\u27s perfonnance. Furthennore, All significantly moderated the relationship between FG and OPP, while it did not significantly moderate the relationship between FG and DMP. This indicates the complexity of the role AI integration plays in SMEs. The paper concludes with recommendations for future research, as well as guidance for practitioners regarding how SMEs can improve their decision-making capabilities and performance using AI
Security Vulnerabilities of a Field Programmable Gate Array
Full text linkThe primary goal of this research is to understand and exploit the security vulnerabilities of a Field Programmable Gate Array (FPGA), at the bitstream level. This paper is working to successfully show that an FPGA can be altered via the bitstream file, and a Trojan can be inserted into the device. Once a Trojan is successfully inserted into the FPGA and activated through a certain input value, a Siamese Neural Network (SNN) will be used to test the effectiveness of Trojan detection. Followed by recording the successful flag rate to detect Trojans, which will be averaged to determine the accuracy score. Trojans are malicious alterations to the original functionality of a device, posing significant risks to data integrity, system functionality, and can even compromise critical infrastructures. This research will address two main questions: How easy it is to manipulate an FPGA at the bitstream level, by altering bitstream files. Having code that once a certain input is entered turns the encryption algorithms off and on, allowing an attacker to see keys and unencrypted data. Secondly, will SNNs flag data when Trojans are present and will it offer high detection rates. Preliminary findings indicate that SNNs offer real-time detection capabilities with high accuracy rates but require a significant amount of training data. The results of this study will contribute to the development of more robust and reliable security measures within FPGA boards. In addition, better protection against cybersecurity threats in various applications, and ways to mitigate these attacks.https://jagworks.southalabama.edu/southalabama-shgrf-posters/1017/thumbnail.jp
Heterogenous Gross Device Deep Learning Power Analysis Attack
No full textDeep learning-based side-channel attacks have shown to outperform non-deep learning-based approaches in certain categories within the side-channel analysis field. One of these categories is cryptographic cross-device side-channel attacks. In these attacks, one or more devices are used to construct a model from side channel data, and that model is used to recover a cryptographic key on a physically different device. Heterogeneous cross-device attacks are a type of cross-device attack where the side channel data used to construct a model is from a device that contains a different processor from a different manufacturer compared to the attack device. In this work, we train a convolutional neural network (CNN) and a multilayer perceptron (MLP) with power consumption data from a Riscure Pinata development board and attack a heterogeneous field programmable gate array (FPGA) device and a similar STM32F device that contains countermeasures. All devices were performing Advanced Encryption Standard 128 bit key (AES-128) with fixed key implementation. Unfortunately, we were unable to successfully recover the key from either attack devices using the MLP or CNN models. However, the models did reduce the key search space and the MLP model reduced the key search space more than the CNN model.https://jagworks.southalabama.edu/southalabama-shgrf-posters/1004/thumbnail.jp
Improving Research Software Engineering in Mathematics
Full text linkResearch Software Engineering is critical to modern mathematical research, enabling the creation, maintenance, and dissemination of computational tools that bridge theory and practice. However, the field faces systemic challenges, including insufficient funding, lack of institutional recognition, and gaps in training and infrastructure. This thesis investigates these challenges through two approaches: (1) a comparative survey study focused on mathematicians and (2) hands-on contributions to an open-source research software project.
The Improving Research Software Engineering in Mathematics survey, conducted from September 2024 to January 2025, adapts the survey framework developed by Carver et al. in A survey of the state of the practice for research software in the United States to analyze research software engineering practices among mathematicians.
The thesis details contributions to pi-Base, a community-driven database of topological counterexamples. Software contributions include a dynamic citation button and an External Markdown feature, both of which were developed through collaborative workflows adhering to continuous integration and continuous deployment (CI/CD) practices. These contributions emphasize the importance of community development efforts in sustaining research software
Linear Dicyclopentadiene Copolymers: Synthesis and Thermal Analysis
Full text linkTo retain the existing advantages of L-pDCPD and to enhance its material properties, an incorporation of functionalized comonomers is being investigated. It is hypothesized that such functionalization will allow for higher theoretical molecular weights, increased oxidative resistance, and more durable thermal properties following air exposure. The use of heteroatoms in exemplary oxa-norbornene-based compounds is thought to strengthen intermolecular interactions and thus improve the structural stability of L-pDCPD. To better understand the polymerization initiated by GC3 and to identify potential ROCM processes, IR-active nitrile and hydroxyl groups were specifically incorporated. For this study, the comonomers were synthesized from exo-3,6-epoxy- 1,2,3,6 tetrahydrophthalic anhydride with either 4-aminobenzonitrile or 4-aminophenol.https://jagworks.southalabama.edu/honors_college_posters/1040/thumbnail.jp
Beyond Traditional Pathways: Addressing Challenges Facing Adult Learners
No full textSuccessful institutions of higher education are continually seeking methods to improve their schools and keep pace with the changing demands of the academic sphere and current job market. With a potential enrollment cliff arising, it is particularly necessary for universities and colleges to find ways to overcome the challenges associated with a possible declining number of traditional—-aged students participating in higher education. A strategically clear solution is to explore the rapidly growing population of adult learners.
This paper will examine the complex motivations, challenges, educational experiences, and perseverance displayed by these students as they work to succeed from admission to graduation. Beginning with insights from my personal background, and building on the results of an informal survey, collection of historical data, demographic statistics and trends, and scholarly research, this thesis demonstrates how adult learners differ from traditional aged students, particularly as a result of their life circumstances, goals, and academic desires. It traces the growth of adult education from its philosophical foundations to today’s advanced technologies and online learning. In doing so, this paper reviews the increasing need for continued adult education.
The findings of this research highlight the strengths adult learners bring to higher education—such as maturity, resilience, and real-world experiences; as well as the barriers that they face, including financial concerns, time poverty, and feelings of isolation. Ultimately, this paper argues that although this demographic is large and the needs required for adult learner success are significant, making the challenge seem daunting, there remains substantial action that universities and colleges can and should take to support not only adult students themselves, but also encourage the growth and success of their own facility. Institutions must adopt intentional, creative, learner-centered strategies to most effectively serve adult learners while recognizing them as essential contributors to the future vitality and success of the school
Learning Without Labels: A Self-Supervised Learning Approach for Anomaly Detection in Control Systmes
Full text linkOil pipelines, water plant systems, and other critical infrastructure are managed and operated by industrial control systems (ICS). These systems safeguard the operations of critical infrastructures, requiring minimal disruption from cyberattacks or malfunctions. The use of anomaly detection methods in control systems (ICS) can reduce system interruptions. However, anomaly detection methods often require annotated data, which may not be available for the control system. Additionally, the datasets used for the control systems do not include sensor outputs and environmental data, resulting in a restricted view of the system. This research investigates how SSL models can be applied to different control system data streams. The study will develop a framework for fusing sensor data, network data, and environmental data. The approach consists of two phases. The first phase includes training SSL models on single data types (e.g., sensor dataset) and an approach for merging multiple data streams. The second phase will combine the SSL model with the aggregated data for anomaly detection. The contributions of this research effort include data fusion framework for control systems and an SSL model trained on the combined dataset without labels for anomaly detection.https://jagworks.southalabama.edu/southalabama-shgrf-posters/1003/thumbnail.jp
Influence of Hearing Loss on Self-reported Hearing Ability and Listening Fatigue in Persons with Aphasia
Full text linkAphasia is an acquired disorder, often caused by stroke, that results in language deficits. It can greatly impact a person\u27s ability to communicate with others. Cognitive deficits may accompany aphasia. Additionally, hearing loss is a disability likely to co-occur with aphasia, as they both most commonly impact the same age group. When a person has hearing loss, the quality of the auditory signal is degraded when it reaches the brain. The poor signal quality, compounded with linguistic and possible cognitive deficits that are associated with aphasia, makes it more difficult for persons with aphasia (PWA) to comprehend speech.
For all people, listening takes effort and attention. The Framework for Understanding Effortful Listening (FUEL) model is an adaptation of Kahneman’s Capacity Model for Attention that helps explain the effort required to listen effectively. Increased effort depletes cognitive resources and leads to listening fatigue. Listening fatigue, when compounded with aphasia and hearing impairment, can exacerbate difficulties comprehending speech. This study examined self-reported listening ability and listening fatigue in people with aphasia and hearing loss.https://jagworks.southalabama.edu/honors_college_posters/1034/thumbnail.jp
False Narratives, Real Consequences
Full text linkSocial media is an increasingly significant tool in modern cyber warfare, capable of rapidly shaping public opinion. The swift dissemination of information complicates efforts to distinguish fact from fiction [1]. During public health crises, healthcare professionals use these platforms to share updates, yet their credible content must contend with false or deliberately misleading narratives [2]. This environment creates an opportunity for cyberattacks through social media influence campaigns [3]. While disinformation\u27s role in political interference has been widely studied, its potential to destabilize healthcare remains largely unexplored. Prior research primarily focuses on how vaccine misinformation affects the general public [4]. This study instead proposes an investigation into the mechanisms, impacts, and broader consequences of disinformation as a method of cyberattack targeting healthcare professionals. Utilizing a three-paper dissertation model, this research will assess disinformation at multiple levels. The first study will analyze social media narratives from the COVID-19 pandemic using natural language processing and sentiment analysis to identify coordinated disinformation campaigns. The second will examine how exposure to misleading health narratives influences nurses’ perceptions, professional judgment, and susceptibility to burnout. The third will simulate a large-scale disinformation attack on a hospital system, modeling its potential to disrupt trust and operational stability. Drawing from prior disinformation campaigns, particularly those during U.S. elections [5][6][7], this study explores whether similar tactics have been or could be used against healthcare. The findings will contribute to cybersecurity, health informatics, and public health preparedness by identifying vulnerabilities that would help safeguard healthcare infrastructure from future information warfare threats. REFERENCES
[1] Hussain, M., & Soomro, T. R. (2023). Social Media: An Exploratory Study of Information, Misinformation, Disinformation, and Malinformation. Applied Computer Systems, 28(1), 13–20. [2] Di Domenico, G., Nunan, D., & Pitardi, V. (2022). Marketplaces of Misinformation: A Study of How Vaccine Misinformation Is Legitimized on Social Media. Journal of Public Policy & Marketing, 41(4), 319–335.https://jagworks.southalabama.edu/southalabama-shgrf-posters/1028/thumbnail.jp
Establishing a Framework for Evaluating Machine Learning Performance and Security across Computational Ecosystems
Full text linkThe rapid evolution of computational ecosystems—ranging from embedded systems and cloud platforms to hybrid and quantum architectures—has introduced new challenges in deploying machine learning (ML) applications. While cloud computing offers scalability, it comes with increased latency and security risks, whereas edge computing, such as FPGA-based systems, provides real-time processing with constrained resources. Hybrid and quantum ecosystems further complicate decision-making, requiring careful trade-offs between performance and security. This research seeks to establish a framework for evaluating ML performance and security risks across these ecosystems, forming the foundation of the Computational Performance And Security System (COMPASS) decision-support tool. The study will systematically investigate key performance indicators—including latency, energy efficiency, and processing power—alongside security concerns such as data privacy, attack vulnerabilities, and system resilience. At this stage, the research focuses on gathering background information, identifying existing gaps, and defining a comparative methodology for analyzing ML deployment trade-offs. The poster will present a literature review, conceptual models, and initial considerations for structuring the COMPASS framework. By addressing these foundational aspects, this work aims to provide a structured approach to optimizing ML performance and security across diverse computing environments.https://jagworks.southalabama.edu/southalabama-shgrf-posters/1015/thumbnail.jp