SACM - United States of America
Permanent URI for this collectionhttps://drepo.sdl.edu.sa/handle/20.500.14154/9668
Browse
6 results
Search Results
Item Restricted ADAPTIVE INTRUSION DETECTION SYSTEM FOR THE INTERNET OF MEDICAL THINGS (IOMT): ENHANCING SECURITY THROUGH IMPROVED MUTUAL INFORMATION FEATURE SELECTION AND META-LEARNING(Towson University, 2024-12) Alalhareth, Mousa; Hong, SungchulThe Internet of Medical Things (IoMT) has revolutionized healthcare by enabling continuous patient monitoring and diagnostics but also introduces significant cybersecurity risks. IoMT devices are vulnerable to cyber-attacks that threaten patient data and safety. To address these challenges, Intrusion Detection Systems (IDS) using machine learning algorithms have been introduced. However, the high data dimensionality in IoMT environments often leads to overfitting and reduced detection accuracy. This dissertation presents several methodologies to enhance IDS performance in IoMT. First, the Logistic Redundancy Coefficient Gradual Upweighting Mutual Information Feature Selection (LRGU-MIFS) method is introduced to balance the trade-off between relevance and redundancy, while improving redundancy estimation in cases of data sparsity. This method achieves 95% accuracy, surpassing the 92% reported in related studies. Second, a fuzzy-based self-tuning Long Short-Term Memory (LSTM) IDS model is proposed, which dynamically adjusts training epochs and uses early stopping to prevent overfitting and underfitting. This model achieves 97% accuracy, a 10% false positive rate, and a 94% detection rate, outperforming prior models that reported 95% accuracy, a 12% false positive rate, and a 93% detection rate. Finally, a performance-driven meta-learning technique for ensemble learning is introduced. This technique dynamically adjusts classifier voting weights based on factors such as accuracy, loss, and prediction confidence levels. As a result, this method achieves 98% accuracy, a 97% detection rate, and a 99% F1 score, while reducing the false positive rate to 10%, surpassing previous results of 97% accuracy, a 93% detection rate, a 97% F1 score, and an 11% false positive rate. These contributions significantly enhance IDS effectiveness in IoMT, providing stronger protection for sensitive medical data and improving the security and reliability of healthcare networks.18 0Item Restricted READINESS ASSESSMENT OF SAUDI NON-PROFIT ORGANIZATIONS: TOWARD CYBERSECURITY AND DATA PROTECTION(Marymount University, 2024) Asiry, Yasser; V Mbaziira, AlexThe increase in cyberattacks in recent years has negatively affected the overall performance of organizations worldwide, leaving them to face significant challenges in improving and strengthening their cybersecurity mechanisms. Non-profit organizations (NPOs) relied heavily on the use of information technology (IT) in their daily operations and strategic initiatives; thus, ensuring the confidentiality, integrity, and availability of sensitive information and the robustness of their systems and policies was crucial. Unfortunately, few studies analyzed the factors that affected the cybersecurity readiness of Saudi NPOs from a holistic point of view. Therefore, this study aimed to assess the readiness of Saudi NPOs in terms of cybersecurity and data protection by examining a comprehensive set of organizational, environmental, and technological factors. The findings of this study were also expected to reveal the positive effects of cybersecurity readiness on organizational performance. This study was likely to encourage governments and decision-makers to improve the cybersecurity readiness of NPOs by introducing cybersecurity initiatives involving security policy interventions, the use of secure resources, and alignment with security regulations. This study was expected to be generalizable to all types of organizations (e.g., profit, non-profit, governmental, non-governmental, etc.) in multiple countries.25 0Item Restricted THE IMPACT OF POPULATION DENSITY AND CYBERSECURITY CHALLENGES IN SMART CITY CREATION(ProQuest, 2024-01-05) Bafail, Ghayda Abdullah; Schaeffer, DonnaWhile the population is growing at a rapid rate worldwide, many people are moving from rural areas to cities when general economic conditions change, looking for good opportunities, better jobs, education, easy life, and better infrastructure. The majority of the population is expected to live in smart cities over the next thirty years. This research contains two parts: quantitative and qualitative. The quantitative part measures the impact of population density on smart-city creation, analyzing thirty-nine countries that have invested in information, communication, and technology (ICT) and ICT goods export for twenty-one years to assess the relationship between population density and smart-city development. The qualitative part briefly discusses the collision of policy, privacy, and ethics in smart cities, which are a top priority in building and developing the smart city, and the main issues policymakers should address when designing smart cities with respect to cybersecurity issues.27 0Item Restricted IDENTIFICATION OF BEHAVIORAL INDICATORS IN MALICIOUS INSIDERS' THREATS IN CYBERSECURITY. A SURVEY QUANTITATIVE STUDY(Saudi Digital Library, 2023-12) Alanazi, Haifa; Liu, MichelleThis dissertation explores the critical issue of insider cybersecurity risks in the context of the oil and gas industry. The paper explores many facets of this phenomena, focusing on the factors that drive the personality traits, possibilities, and capabilities of these insiders as the industry faces an increasing threat from hostile insiders. This study's main goal is to provide a thorough examination of insider threats in the oil and gas industry with a focus on comprehending the underlying motivations and behavioral indications. The author aims to offer insightful information that can improve the cybersecurity practices used by the sector. The lack of focus on insider threats in the oil and gas industry is one obvious gap in the current body of literature. While external cybersecurity threats have been the subject of substantial research, insider threats are still largely unexplored. The dissertation examines the area of insider cybersecurity threats in the oil and gas industry by delving into four key research areas. These inquiries include a look at the motives that drive nefarious insiders within the industry, a look at how personality traits affect insider threats, a look at the circumstances that make it possible for such threats to happen, and a look at how insiders' skills affect the industry's overall cybersecurity risks. A quantitative approach was used to study these issues, and 95 oil and gas sector participants were polled. The questionnaires used a Likert scale. With the aid of the proper statistical software, descriptive and regression analysis of the data were performed. The results emphasize the critical part that motivations, personality, opportunity, and capabilities play in determining malevolent insider behavior. Notably, motives—such as unintended mistakes—were discovered to be important contributors to insider threats. According to the research, effective cybersecurity must include preventative measures, personnel training, and ongoing security assessments. This thesis offers the oil and gas sector insights to strengthen its cybersecurity defenses against insider threats, which has practical ramifications. It highlights the need for systematic preventive actions, an interdisciplinary strategy, and increased employee awareness to reduce the hazards brought on by hostile insiders. Additionally, it emphasizes how cybersecurity tactics are constantly altering in response to the shifting insider threat picture within the industry.44 0Item Restricted Smart Home Cybersecurity Challenges: An Assessment of End-User Knowledge and a Training Solution to Mitigate these Challenges.(Saudi Digital Library, 2023-11-22) Nusair, Ali; Chipidza, WallaceAs the digital revolution unfolds, individuals are increasingly transforming their traditional homes into smart homes, adopting semi- and fully automated smart devices. This transformative shift, fueled by advancements in information technology, presents vast social and economic opportunities. Despite the burgeoning number of smart devices in the market, a surge in smart home adoption has concurrently given rise to profound security challenges. Predominantly, end-users, often possessing rudimentary knowledge of associated risks, remain vulnerable to breaches of their privacy and security. Given that smart devices, interconnected and internet-enabled, relay substantial data, they are attractive targets for hackers. One fundamental reason for these challenges is the end-users' lack of requisite knowledge to safeguard their smart homes. To address these challenges, there's a pressing need for effective knowledge dissemination. This dissertation introduces two artifacts: a training framework detailing smart home vulnerabilities and best practices for cybersecurity, and an application named "Smart Home Security App". This application prompts users to update their passwords biannually and continuously monitors for potential security breaches. Drawing from an extensive literature review, the two artifacts were developed. To evaluate the framework's effectiveness, a set of 34 survey questions was crafted, reflecting key cybersecurity knowledge areas. Fifteen participants, after providing written consent, responded to these questions. Their initial responses informed the development of the first artifact, and post-training, the same questions were administered. Notably, there was a marked enhancement in the participants' understanding of smart home security post-training. Leveraging the Design Science Research methodology, the artifact's efficacy as a consumer training tool was assessed. Keywords: Smart home, IoT, vulnerabilities, smart devices, cybersecurity, hacking, social engineering, identity theft, Smart Home Security App.28 0Item Restricted Machine Learning (ML) Technologies(2022-11-14) alshammari, Atiah; Seferaj, GentianaThis research paper aims to analyze ML technologies primarily used in physical security, their tangible results, and the impact of these trends on the future of physical security.57 0