The large-scale application of IoT technology in the medical industry also brings the advantage of continuous monitoring of patients and timely detection of heart diseases; however, it also results in challenges like ensuring data privacy, resources, and management of medical records. The paper presents a secure hybrid machine learning framework incorporated with IoT for diagnosing heart diseases at their infant stage, mostly meant for healthcare settings with low resources. The proposed protocol integrates lightweight encryption to secure medical data during early-stage diagnosis and data transmission in IoT-based healthcare systems combined with machine learning models, while nature-inspired optimization improves feature selection, classifier tuning, accuracy, and convergence. A variety of cardiac datasets are extensively used for the system's proposal, and it is evaluated under inconsistent and noisy IoT conditions to determine its generalization capacity, robustness, and interpretability, which are assessed through the deployment of explainable artificial intelligence techniques. The framework is also analyzed regarding scalability, computational efficiency, and security - performance trade-offs to verify its suitability for real-time deployment in large-scale IoT healthcare systems. The findings of the study indicate that the suggested method has a well-balanced integration of security, accuracy, interpretability, and efficiency, thereby giving a practical and reliable solution for intelligent IoT-based healthcare diagnostics.
Challenges faced by students and professionals with visual impairment in reading and understanding digital content have always been a great area of concern, particularly in subjects like mathematics and science due to their high dependency on special symbols. On one hand, useful tools are limited by complex formats; on the other hand, most available tools are directional dependent as standard textual information does not support an effective tactile form of knowledge. This research therefore, developed Optical Braille Recognition Methodology (OBRM) for transforming printed documents into Braille language files. The proposed methodology integrates image segmentation with multistage AI processing to enhance recognition of textual content and interpretation of mathematical and special symbols. A novel key-element quantification strategy reduces overall complexity and minimizes memory usage. Multi-format document acquisition followed pre-processing steps to enhance quality and remove noises before starting OCR processing that again used additional Neural Network backend for structured parsing which finally ended the tactile symbol generation stage, where new 3-bit encoding implemented readable form creation in a BRF extension file output. Even the semantic structure organized text making similar word or symbol placed together assists easier understanding during the later tactile reading learning process. A user-experience survey involving target users was organized and carried out on the system as a beta test.
This paper explores cybersecurity compliance behavior in organizations, focusing on the interplay of individual and organizational factors. Employing Protection Motivation Theory (PMT), the Theory of Planned Behavior (TPB), and Organizational Culture Theory (OCT), the study examines how employees' perceptions of threat severity, vulnerability, response efficacy, and response cost influence their compliance intentions. It also assesses the impact of organizational cybersecurity efforts and employee security awareness. The research, guided by insights from Rogers, Ajzen, and Schein, utilizes a quantitative methodology with Partial Least Squares Structural Equation Modeling (PLS-SEM) and Necessary Condition Analysis (NCA) to analyze complex relationships between these factors. The findings highlight the significant role of individual decision-making styles, organizational culture, and the effectiveness of security technologies in shaping cybersecurity compliance. This study contributes to the literature by providing a comprehensive understanding of cybersecurity compliance behavior, offering valuable insights for organizations to develop effective strategies and enhance their cybersecurity posture in an increasingly digital world.
With the fast and continuous increase in frequency and sophistication, securing significant data - such as passwords, PINs, and financial credentials - has become a critical issue. This paper presents a block-level sensitive data classification framework that detects and separates blocks of data encapsulating high strategic data or sensitive information. Employing an integrated approach, entropy analysis, and deep learning-based content detection, sensitive blocks are flagged in real-time and secured using an additional encryption layer. This significantly improves security mechanisms and mitigates the risk of unauthorized access and data breaches. The proposed architecture was evaluated using real-world commercial financial datasets and resulted in a mean threat mitigation accuracy of 96.3%, outperforming traditional full-disk encryption methods. This research provides an effective framework for banking and enterprise systems, offering a proactive encryption layer against advanced and sophisticated cyber-attacks.
Attack Detection (AD) in healthcare-based wireless sensor networks (H-WSNs) presents major challenges due to low processing speeds, limited storage, poor detection rates, long deployment times, restricted communication ranges, and reduced energy efficiency. These limitations hinder effective implementation and expose networks to security threats. The complexity of existing methods, especially the curse of dimensionality, further reduces detection efficiency and accuracy. This research addresses these challenges by optimizing storage, computational time, and detection accuracy through a three-phase approach: pre-processing, dimensionality reduction, and classification. First, raw patient health data is enhanced using the Guided Box Filtering (GBF) method. Next, dimensionality reduction is performed using Principal Component Analysis (PCA) with Lotus Effect Optimization (LEO), termed QLPAFMT-LEO. This hybrid method improves parameter interpretation, reduces computational time, and minimizes space complexity. Finally, the Stereoscopic Scalable Quantum Convolutional Neural Networks (SSQCNN) model is applied to classify normal, black hole, and grayhole attacks. To enhance accuracy, the Gooseneck Barnacle Optimization Algorithm (GBOA) optimizes SSQCNN parameters. The proposed SSQCNN-GBOA model is validated using the NS2.34 network simulator, demonstrating superior AD, optimized computational efficiency, and reduced energy consumption. This approach significantly enhances H-WSN security and operational performance, making it a robust solution for real-world deployment.
Cyber-physical systems (CPS) are described as the latest invention of complicated networks, which combine both the physical and cyber worlds. Because of their strong coupling nature, it is considered that a few elements may fail to be recognized within a particular duration. Deep learning models offer effective approximate actions and representation-based learning abilities, permitting conventional techniques to understand the most effective guidelines for sequential decision-making within complicated surroundings, particularly while handling high-dimensional inputs and huge-scale issues. Therefore, a new strategy is recommended to identify cyber anomalies and attacks within the CPS using adaptive deep learning. Initially, the required input data are fetched from various benchmark datasets, and these gathered data are fed into the feature-extraction procedure. Here, the Stacked Autoencoder is utilized for extracting the deep features. Further, the resultant features are given as input to the Adaptive Efficient Capsule Network to determine cyber anomalies and attacks in the CPS. Here, a Modified Random Number-based Garter Snake Optimization Algorithm is used for optimally tuning the parameters in the CapsNet. Finally, detailed experiments are carried out for the developed cyber-anomaly and attack-detection approach to guarantee the effectiveness of the implemented mechanism over traditional techniques.
The widespread application of digital technologies in urban green landscape design has brought new opportunities while introducing cybersecurity challenges. Frequent occurrences of cyber attacks, data breaches, and privacy infringements severely threaten the stability and data security of urban green landscape systems. This study analyzes current applications in spatial thermal energy regulation and urban landscape governance, proposing strategies including thermal energy scheduling coordination, energy storage, optimized scheduling models, fault-tolerant scheduling strategies, and thermal energy regulation simulations. An AI-based urban landscape governance system is constructed, along with an ecological safety assessment methodology for landscapes, along with evaluations of its current status and system architecture. The research particularly focuses on cybersecurity issues, ensuring data security and operational stability through encrypted communication protocols, firewall configurations, vulnerability scanning, and real-time monitoring and early warning of cyber attacks using AI technology. The findings reveal that the AI-powered spatial thermal energy regulation system significantly enhances the stability and ecological security of urban green landscapes. Optimized scheduling effectively reduces urban heat island effects, strengthens internal ecological connectivity, and substantially improves the effectiveness of urban green landscape governance. In terms of cybersecurity, multi-layered protective measures successfully resist various cyberattacks, safeguarding data integrity and operational stability. Practical evidence demonstrates that AI systems integrated with cybersecurity measures exhibit significant advantages in urban green landscape governance.
Cybersecurity in the Internet of Things (IoT) has become the fastest developing technology, with a significant impact on social life and commercial environments. The number of threats is increasing every day, and attacks are becoming more numerous and complicated. However, existing techniques lead to information loss and slow learning during feature extraction, making the process difficult and increasing computational complexity. Hence, a novel Cascaded Denoising Autoencoder (DAE)-based optimized Back Propagation (BP) algorithm is introduced to enhance feature extraction efficiency, reduce noise during feature overlap, accelerate training speed, and improve accuracy, while minimizing computational complexity. The existing techniques suffer from the multiclass classification problem which leads to lower performance in classification and detection. Hence, a novel Softmax-based Multilayer Perceptron Neural Network with Support Vector Machine (MLPNN-SVM) model is introduced, which combines the neural network with a classifier to enhance the accuracy of classification and detection, reducing the false positive rate and computational cost. The proposed model uses the Network Security Laboratory-KDD (NSL-KDD) dataset for evaluating intrusion detection systems. This dataset contains various cyber-attacks and normal traffic data. As a result, the MLPNN-SVM model detects and classifies cyber-attacks, reducing the false positive rate and computational cost, leading to an achievement of an accuracy of 99.23%, precision of 99.24%, and recall of 99.22%, significantly improving overall system performance.
A new framework was developed for enhancing QoS through optimal resource allocation and base station optimization. The energy efficiency of the system under the QoS constraints is enhanced by the resource allocation strategy. This is achieved by the Adaptive and Attention-based Deep Deterministic Policy Gradient (AA-DDPG) method for energy-efficient resource allocation and base station optimization in the 6G network. In developed AA-DDPG, attributes are used as the input, which helps to offer the resource allocated outcomes by tuning the parameters of AA-DDPG through Random Number Updated Bobcat Optimization Algorithm (RNU-BOA). Tuning the parameters of AA-DDPG supports to obtained optimal resource allocation and base station, which helps to enhance the energy efficiency of the network. Then, extensive simulations are carried out and they demonstrate that the suggested approach ensures sustainable operation and provides an impressive outcome than others. The developed resource allocation and base station optimization model attained higher energy efficiency as 88.78%, security rate as 83.68%, resource utilization rate as 48.3%, and link utilization rate as 44.1% better than the classical schemes like HLFLM, MAAC, MARL-DDQN, and DQN-GAN, respectively. Hence, the validation outcomes displayed that the proposed model is better to offer better outcomes without any errors.
Multiprotocol Label Switching (MPLS) networks provide efficient traffic engineering and QoS support, but they remain exposed to volumetric denial-of-service threats that can degrade availability and service quality. This paper proposes a lightweight, controller resident intrusion detection and prevention system (IDPS) for wireless MPLS networks using Software-Defined Networking (SDN). The proposed application combines two anomaly features packet rate (PPS) and packet size (MTU/oversize packets) to identify UDP flooding behavior and to trigger automated mitigation. Upon repeated violations, the SDN controller installs OpenFlow drop rules at the ingress switch to blacklist offending sources and reduce control plane load after rule deployment. The solution is implemented in OMNeT++/INET with an MPLS - SDN topology and evaluated under a high-rate UDP flood scenario ($ \ge 10,000$>= 10,000 pps with oversize packets). Performance is assessed using detection rate, end-to-end delay, packet delay variation, and energy consumption. The results show that the proposed IDPS improves traffic isolation and attack mitigation while maintaining acceptable QoS and energy efficiency.
In today's globalized and interconnected world, forming partnerships is essential for most organizations to succeed. The partners in such an Interorganizational Relation (IOR) can support each other in the production of their products, offer joint ser-vices, exchange resources or work together on innovation projects. Beyond common business goals, these IORs offer the opportunity to engage in joint Information Security Management (ISM), an angle that has received little attention so far. This conceptual research investigates which of the common ISM processes, according to ISO/IEC 270022, would be eligible for an interorganizational approach. To this end, a partnership model is introduced that defines three types of IORs: Supportive Relationship, Cooperative Relationship or Collaborative Relationship. Building upon this model, the suitability of IORs for each ISM process is evaluated by analyzing three generic forms of cooperation: exchanging data, pooling resources and jointly execute (an activity). This article presents a new research perspective to advance the field of joint ISM and contributes a novel partnership model that classifies IORs into generic types defined by their core characteristics and a currently missing overview of ISM processes showing to what extent they can potentially benefit from an interorganizational approach.
Small and medium-sized enterprises (SMEs) encounter cyber security risks, yet the factors underlying variation in these risks remain unclear. This study examined whether 1) cyber security controls, 2) awareness, knowledge, attitude, culture and 3) support routes vary according to SME characteristics (size, type, maturity and sector). It also explored the factors that shape SMEs' decisions to access resources that help reduce cyber security risk. A mixed-methods design combined a survey of 374 participants and interviews with 12 SMEs. ANOVAs analyzed differences across organizational categories, and thematic analysis was applied to qualitative data. The study shows that many SMEs in the sample have inadequate security controls and limited awareness, knowledge and capability in cyber security. Qualitative insights show that SMEs may underestimate risk, face competing priorities and are unsure where to find support. These findings highlight the need for practical, accessible support to help SMEs implement effective cyber security.
The organizations' structural units are designed to combat computer attacks on their Information and Telecommunication Networks (ITCNs). To be effective, these units should include Network Security Centers (NSCs) that focus on specialized functions and possess capabilities for ITCNs'Network Security Management (NSM). This should be achieved through a comprehensive and systematic approach. NSCs must demonstrate a high level of maturity that aligns with the information security (IS) requirements applicable to their respective organizations. There are various approaches for assessing the maturity level (ML) of Security Operations Centers (SOCs), which serve as the first generation of NSCs. However, there is no unified assessment methodology available. It would be useful for identifying ways to improve NSCs and would provide a consistent framework for authorized bodies when conducting assessments. The article's goals are twofold: first, to critically analyze existing models for assessing the SOCs' MLs to understand their advantages and disadvantages; and second, to develop requirements for a unified model for assessing NSC maturity, drawing from best practices while addressing identified shortcomings. These requirements are grouped into five categories: maturity assessment methodology, directions, objects, methods, and MLs. The model and the corresponding methodology are currently in their final stages.
Zero-day attacks exploit vulnerabilities that have not been detected before and therefore do not pose a problem for existing signature-based intrusion detection systems, as they lack established attack patterns. Although machine learning-based detection has improved, current methods are not flexible enough to handle new threats and have high false-positive rates. In this paper, a Deep Reinforcement Learning (DRL) framework is proposed to design a zero-day attack detection system as a Markov Decision Process (MDP) that supports adaptive learning without using any attack signatures. We apply and compare three DRL algorithms, Deep Q-Network (DQN), Proximal Policy Optimization (PPO), and Advantage Actor-Critic (A2C), with a new feature engineering method that combines Principal Component Analysis with Information Gain selection. The framework is tested on various benchmark datasets (NSL-KDD, CICIDS2017, CIC-AndMal2017) and a specially created dataset of zero-day attacks in the context of the present research. Most experimental findings show that the DQN model attains 91.7% accuracy and 83.4% detection rate over previously unseen attacks, 14.7% and 8.9% better than traditional machine learning and deep learning baselines, respectively. A study of ablation indicates that the exploration strategy plays a critical role in zero-day detection, with its removal resulting in a 10.2% reduction in detection rates. The suggested framework offers greater flexibility against different types of attacks while maintaining a lower false-positive rate (8.2%) than traditional methods. The work contributes to the development of cybersecurity defense functions by demonstrating that DRL is a useful paradigm for detecting unknown threats in dynamic network settings.
The underrepresentation of women in cybersecurity, along with critical skills shortages, persists. This study explores the challenges and barriers women face in the cybersecurity profession in South Africa and proposes interventions to promote inclusivity. A qualitative study was conducted among female cybersecurity professionals in South Africa. Thematic analysis was performed using Atlas.ti. The results identified challenges emanating from an organizational level, educational barriers, societal stereotypes and resource constraints. Women in a male-dominated profession often have a dual family responsibility role and face challenges in organizational support and equity. Societal biases and stereotypes challenge females from entering STEM careers, while inadequate resources and infrastructure affect the talent pipeline. Interventions include gender equity, diversity and inclusion training, as well as cybersecurity, awareness and education. Changes on an organizational level in policies, processes, and culture, as well as educational transformation supported through government strategy, are recommended. The study offers recommendations for organizations, the government, policymakers and educators to foster an equitable environment and reduce barriers. Including more women in cybersecurity can help bridge the skills gap and bring diverse skills, leading to more creative and innovative solutions that aid organizations in becoming more cyber resilient.
Over the past decade, Intelligent Transportation Systems (ITS) have experienced remarkable transformation, fueled by advancements in data analytics, sensor networks, and machine learning. One of the most critical aspects of ITS is the detection of anomalies in traffic flow, which plays a key role in ensuring road safety, optimizing traffic management, and responding promptly to unexpected events such as accidents, road blockages, or sensor malfunctions. This research paper explores the application of time series analysis techniques for anomaly detection in ITS, with a particular focus on three prominent approaches: Dynamic Time Warping (DTW), Matrix Profile using the STOMP algorithm, and the Prophet forecasting model. The Prophet model demonstrated the highest F1-score of 0.88, outperforming DTW (0.81) and Matrix Profile (0.75), showcasing its superior ability to handle seasonality and trend components in traffic data for accurate anomaly detection. Additionally, the Prophet model is investigated for its strength in handling seasonality and trend components in traffic data while offering interpretable anomaly detection. The study provides a comparative analysis of these methods based on accuracy, scalability, interpretability, and suitability for real-time deployment. This paper concludes by discussing future directions in ITS anomaly detection, including the integration of hybrid models and real-time adaptive frameworks to enhance predictive capabilities and operational resilience.
