Network Intrusion Research Articles

A Review on Cloud Security: Challenges, Solutions, and Innovations

Cloud computing has become one of the fastest growing and most advance technological evolutions providing companies and individuals the means to scale resources, as well as being flexible and saving cost. While these advantages come with high security risks such as data breaches, unauthorized access, and service disruptions respectively. We review and compare five recent studies that cover aspects of cloud security from encryption, to threat detection, to multi-cloud challenges, to quantum resistant solutions. The papers offer useful information about the emerging security techniques and frameworks. The methodologies, results, limitations and areas for future discussion are presented in a comparative analysis

Unveiling anomalies: transformative insights from transformer-based autoencoder models

Nowadays, Intrusion Detection Systems (IDS) play a critical role in safeguarding networks by identifying and mitigating unauthorized access and malicious activities, yet face challenges in accurately detecting sophisticated threats while minimizing false positives. In this paper, we present a comprehensive analysis of a Transformer-based Autoencoder model for network intrusion detection using the NSL-KDD dataset. We evaluate the model's anomaly detection performance based on key metrics such as Mean Squared Error (MSE), Precision-Recall Area Under Curve (AUC), and Root Mean Squared Error (RMSE) on the test set. Our findings reveal promising results, with the model achieving low MSE, high AUC, and relatively low RMSE values, indicating its effectiveness in detecting anomalies. Additionally, we provide an in-depth examination of the model architecture, highlighting the role of encoder and decoder layers, dropout regularization, and optimization techniques such as Adam optimizer. Our analysis sheds light on the efficacy of Transformer-based Autoencoder models for network intrusion detection tasks, offering insights into their architectural design and performance evaluation. The novelty of this study lies in the implementation of a Transformer-based Autoencoder architecture for network intrusion detection, showcasing its superior performance in anomaly detection compared to traditional methods.

Enhancing Security and Accessibility of Personal Records in Nigeria using Blockchain Technology: A Case Study of Modibbo Adama University, Yola

The security and accessibility of personal records have become critical challenges in Nigeria, given the country's reliance on outdated and vulnerable systems for managing sensitive data. This research explores the role of blockchain technology in enhancing the security and accessibility of personal records in Nigeria. Blockchain, known for its decentralised, transparent, and tamper-resistant nature, offers a promising solution to issues such as data breaches, unauthorized access, and inefficient record-keeping. The study investigates how blockchain can provide a more secure and reliable infrastructure for managing personal records, focusing on key attributes such as encryption, immutability, and data transparency. It also examines the potential for improving accessibility, especially in areas where record retrieval is often delayed or compromised by bureaucratic bottlenecks. By evaluating case studies and existing applications of blockchain technology in record management, this research identifies opportunities and challenges for blockchain adoption in Nigeria's public and private sectors. Through a mixed-method approach that includes surveys and interviews with key stakeholders in data management, the study highlights the feasibility of integrating blockchain into Nigeria’s personal records system. The findings suggest that blockchain technology, despite certain challenges such as cost and technological infrastructure, can significantly enhance the security and accessibility of personal records in Nigeria. Recommendations for policy development, technological investment, and further research are also provided.

TinyNIDS: CNN‐Based Network Intrusion Detection System on TinyML Models in 6G Environments

ABSTRACTWith the evolution of network technologies from 5G to 6G, security has become increasingly critical due to the vast number of connected devices and the decentralized nature of network architectures. TinyNIDS addresses these challenges by leveraging a lightweight CNN (TinyCNN) optimized through quantization and pruning, enabling real‐time intrusion detection on resource‐constrained edge devices. This paper presents TinyNIDS, a CNN‐based network intrusion detection system designed for deployment on TinyML models in 6G environments. The system aligns with the demands of 6G networks, including ultra‐low latency, high throughput, and decentralized processing, ensuring robust threat detection without centralized bottlenecks. Extensive experiments demonstrate the superior performance of TinyCNN compared to traditional models, achieving higher accuracy, lower false positive and false negative rates, and reduced latency. The results validate the effectiveness of TinyNIDS as a scalable and efficient solution for real‐time network security in next‐generation networks.

Generating Credibility in Crisis: Will an AI-Scripted Response Be Accepted?

ABSTRACT Increasingly, practitioners are using artificial intelligence (AI) to strategically monitor and respond to crises. However, there is little evidence indicating whether a crisis response, disclosed as AI-scripted, will be accepted by stakeholders and what effects the disclosure may have on message credibility, attribution of responsibility, message acceptance, and organizational reputation. Using Situational Crisis Communication Theory (SCCT), this 2 (type of crisis) x 2 (presence or absence of AI label) online experiment explored how the disclosure of AI-generated content affects post-crisis organizational outcomes. Participants (n = 238) were randomly assigned to a vignette featuring a victim (e.g. shooting) or accidental (e.g. data breach) cluster crisis and a detailed response that either disclosed or did not disclose that the content was AI-generated. Results revealed no effect of disclosure on message credibility or attribution of responsibility. However, message acceptance served as a mediator between message credibility and attribution of responsibility on organizational reputation. This study contributes to SCCT and provides contextual evidence for practitioners who are considering AI for crisis responses. Ethical implications and future directions are discussed.

Cybersecurity strategies for non-profit organizations

The paper explores effective cybersecurity strategies tailored for non-profit organizations, addressing their unique challenges of limited resources and sensitive data handling. It examines the current threat landscape, highlighting common vulnerabilities such as phishing attacks, ransomware, and data breaches. The research presents key cybersecurity strategies, including risk assessment and management, employee training and awareness, implementation of technical controls, data protection and privacy measures, and third-party risk management. Cost-effective solutions are emphasized, focusing on open-source security tools, cloud-based services, and collaborative information-sharing networks. The study concludes that by adopting a holistic approach encompassing people, processes, and technology, non-profits can significantly enhance their security posture and resilience against cyber threats, ensuring the continuity of their critical missions while operating within resource constraints.

Improved convolution neural network integrating attention based deep sparse auto encoder for network intrusion detection

Improved convolution neural network integrating attention based deep sparse auto encoder for network intrusion detection

Development of algorithms for early detection of cyberattacks on networks using machine learning

Critical infrastructure protection and national security are enhanced by the security and reliability of networks. Various types of information circulate on these networks, ranging in classification from open to closed. The consequences of cyberattacks on these networks can be severe, including reputational damage, financial loss, operational disruption and data leakage. Traditional security methods, such as firewalls and anti-virus software, are becoming less effective against modern and ever-changing cyber threats. As a result, powerful network intrusion detection systems (IDS) have become indispensable for proactive detection and mitigation of cyber attacks. Machine learning has become a viable method for creating adaptive intrusion detection tools that can detect new and complex attack patterns. By learning from huge labelled network traffic datasets, ML models can understand the subtle patterns and differentiating features of normal and abnormal or malicious traffic flows. This allows them to detect possible cyber threats and intrusions that traditional signature-based IDSs cannot detect. Extracting discriminative features and training appropriate classification models from such data is a challenging task. In the presented study, we analyse the effectiveness of ML algorithms for detecting cyberattacks, in particular distributed denial of service (DDoS) attacks, in network traffic data. In the presented study, a network attack detection system is developed using ML and deep learning (DL) models and experimented on the CICIDS2017 dataset. The main objectives of the study are to develop a strategy for extracting valuable information from raw network streams; to study the impact of data preparation on the false positive rate; and to conduct a comparative analysis of ML models for cyberattack detection. The main goal of the study is to provide an understanding of the development of a reliable adaptive network intrusion detection system using ML approaches that increase cybersecurity capabilities and protect against future cyberattacks.

Method for raising personnel awareness of information security using the Gophish software application

In today’s cyberspace, where threats are constantly evolving, phishing attacks require special attention. They are one of the most common social engineering methods used to gain access to confidential information by manipulating users. Such attacks can lead to data breaches, financial losses, and reputational risks. One of the key problems is the lack of staff training in recognizing phishing threats. Traditional training methods do not provide adequate interactivity and realism, which reduces the effectiveness of awareness raising. To address this issue, new approaches are needed that model real-life cyberattack scenarios. The Gophish tool allows you to create personalized phishing campaigns that simulate real attacks and analyze user reactions. Its functionality includes creating email templates, sending messages, and collecting data on user interaction. This allows organizations to identify weaknesses, adjust training programs, and conduct additional training. The results show that using Gophish increases staff awareness of social engineering. Interactive simulations contribute to a better understanding of phishing threats and help users recognize the danger in time. Thanks to the analytics collected, management can quickly implement corrective measures. The development of artificial intelligence and machine learning opens up new opportunities for improving such tools. Future solutions will be able to adapt simulations to specific users, making training even more effective.

Blockchain and Machine Learning: A Multidisciplinary Synergistic Approach for Fraud Detection in Finance, Healthcare, and Cybersecurity

The integration of Blockchain and Machine Learning (ML) technologies offers a transformative approach to combating fraud across various sectors, including finance, healthcare, and cybersecurity. Blockchain's decentralized and immutable nature ensures data integrity and transparency, while Machine Learning algorithms enable the detection of intricate fraud patterns through predictive analytics and anomaly detection. This synergistic combination provides a robust mechanism for identifying fraudulent activities in real time, minimizing human error, and optimizing decision-making processes. In the financial sector, Blockchain enhances the security and transparency of transactions, while ML models analyze transaction data to identify unusual patterns that may indicate fraud. In healthcare, Blockchain ensures the secure sharing of medical records, and ML assists in detecting fraudulent claims and potential identity theft. Cybersecurity applications leverage Blockchain for secure communication and data storage, with ML identifying potential threats or vulnerabilities. By combining these two cutting-edge technologies, organizations can strengthen their fraud detection systems, improve trust, and mitigate the risks associated with financial losses, data breaches, and privacy violations. This paper explores the multidisciplinary synergy of Blockchain and ML, illustrating their potential to revolutionize fraud detection mechanisms across multiple domains, providing a comprehensive overview of current advancements, challenges, and future directions for their integration in the fight against fraud.

A Research on the Protection of Personal Information in the Context of Face Recognition Technology

The rapid development of science and technology, coupled with the extensive use of the Internet, has led to the emergence of face recognition technology as a prominent high-tech power, which has become an integrated part of people’s daily life. While facilitating convenience in life, face recognition technology also presents a multitude of legal risks and challenges. The diversity and opacity of the Internet provide an optimal environment for those engaged in illicit activities to exploit its shortcomings for a range of illegal and criminal pursuits. As restrictions on the use of face recognition technology are lifted and its scope of application expands, the risk of data leakage and violations of individual privacy, as well as its potential use as a tool for crimes, continues to rise. China still has deficiencies and vulnerability in the implementation of face recognition technology. For instance, the pertinent legislation and regulatory frameworks governing the face recognition industry and the collection and utilization of personal data remain somewhat ambiguous and incomplete, which may result in the infringement of privacy rights. It would be prudent for China to draw upon the experiences and solutions that have been developed domestically and abroad to improve and tailor the existing legal system and regulations to the country's specific circumstances. This approach would serve to reinforce the protection of personal information, facilitate the more effective utilization of the technology for the benefit of the nation and society, enhance convenience for citizens, and reinforce their sense of security and well-being.

Early Detection of Hard-Coded Secrets in Software Development: A Multi-Method Approach Integrating Static Analysis, Entropy-Based Detection, and Machine Learning

The inadvertent inclusion of hard-coded secrets—such as API keys, passwords, and tokens—within source code poses significant security risks, potentially leading to unauthorized access and data breaches. Recent studies have highlighted the prevalence of this issue; for instance, GitGuardian's 2023 report revealed a 67% increase in detected hard-coded secrets compared to the previous year, with 10 million new secrets discovered in public GitHub commits in 2022. To address this escalating concern, our research proposes a comprehensive, multi-method approach that integrates static code analysis, entropy-based detection, and machine learning techniques to enhance the early identification of hard-coded secrets during the software development lifecycle. By combining these methodologies, the proposed approach improves detection accuracy, reduces false positives, and provides developers with actionable insights to mitigate security vulnerabilities proactively. This integrated strategy not only strengthens the security posture of software applications but also fosters a culture of secure coding practices among development teams. Keywords Shift-Left Security, Secrets Detection, Static Analysis, Entropy-Based Detection, Machine Learning, Hard-Coded Secrets, Secure Development Lifecycle

Encryption and Security Enhanced Middleware SOA Solutions for High-Security Applications

This paper examines the security enhancements for middleware layer that play a crucial role in protecting information. In the digital age, the integration of security measures with software architecture is paramount, especially for industries such as banking, healthcare, and high-security applications. These sectors handle sensitive data that requires robust protection from cyber threats. The Service-Oriented Architecture (SOA) is a widely adopted architectural pattern that facilitates the design of modular, interoperable systems. However, to address the escalating risks associated with cybercrime and data breaches, traditional SOA solutions must evolve by incorporating encryption and other advanced security measures. This paper explores how encryption and security- enhanced middleware enhances the overall security for these high-stakes industries. Keywords: SOA, OIC, Webservices, XML, TLS, SSL, Encryption, Data encryption, Encryption at rest. SAML, SSO, Application Security

Design and analysis of H∞$$ {H}_{\infty } $$ control strategies with dynamic event‐triggered schemes and quantizer in the presence of hybrid attacks

AbstractThis study explores the resilience of networked control systems (NCSs) against hybrid attacks, focusing on enhancing their performance. To mitigate excessive data transmission, the article introduces dynamic event‐triggered schemes and quantizer within the sensor‐to‐observer communication channel. The network environment is susceptible to various attacks, including denial‐of‐service (DoS) attacks and deception attacks. An augmented system model is formulated, and through the application of Lyapunov stability theory, conditions for ensuring the asymptotic stability of the system are established. The collaborative design of quantizer parameters, controller gain, and event‐triggered thresholds is achieved by solving linear matrix inequalities. Simulation‐based theoretical models have effectively demonstrated the reduction of the impact of network intrusions on NCSs, validating the proposed approach.

Data Migration and security in cloud

Data migration and security in the cloud" refers to the process of transferring data from an on-premises system to a cloud environment while prioritizing security measures to protect sensitive information during transit and once stored in the cloud, including implementing strong access controls, encryption, and regular monitoring to prevent data breaches and ensure compliance with relevant regulations. In this time of digital evolution, cloud migration and its security have become essential topics that cannot be avoided. With more companies turning to cloud services for support, the demand for large-scale data storage and data management has also increased. Along with this rising need, the risk of cyber attacks on cloud data has also increased. To better protect yourself from these risks, it’s essential to ensure that you and your organization are up-to-date on the latest security technologies and safety precautions. Keywords : Data Migration, Data Security, cloud, ETL, Data Integration

Security and Compliance in Cloud ERP Systems: A Deep Dive into Workday's Framework

This paper provides a comprehensive analysis of the security and compliance framework within cloud- based ERP systems, with a particular focus on Workday, a leader in the market for enterprise resource planning (ERP) solutions. As businesses increasingly transition to cloud ERP systems, ensuring robust security and regulatory compliance has become critical to safeguarding sensitive data, maintaining operational integrity, and meeting industry-specific regulations. The paper begins by introducing the essential concepts of cloud ERP security, highlighting the challenges of data breaches, data integrity, and access control. It further explores Workday's advanced security infrastructure, including its encryption protocols, secure data storage, and real-time monitoring systems, as well as its adherence to global compliance standards such as GDPR, SOC 2, and ISO 27001. The analysis also emphasizes Workday’s tailored security features for specific industries like finance, healthcare, and government, demonstrating how its platform fosters customer trust and drives adoption across diverse sectors. Looking to the future, the paper discusses emerging trends, including the integration of Artificial Intelligence (AI), Machine Learning (ML), and blockchain technology, which hold potential for enhancing data security and compliance in cloud ERP environments. This deep dive into Workday’s security and compliance framework underscores its role in mitigating risks, ensuring regulatory adherence, and positioning itself as a trusted solution in an evolving cloud landscape. Keywords: AI/ML, Cloud ERP Security, Workday Security Framework, Cloud ERP Compliance, Workday Compliance Standards, Data Security, Workday Data Encryption, Regulatory Compliance, SOC 2 Compliance, Identity and Access Management, GDPR Compliance, Risk Mitigation.

The Effects of Auditor-level Cybersecurity Breaches on Auditor-Client Relationships

Between 2005 and 2019, Big4 auditors experienced fifteen data breaches. In this paper, we examine how client firms respond to an auditor-level data breach and whether auditors adjust their behavior in response. Using auditor-client pair observations, we find a higher likelihood of auditor turnover among clients of breach-affected auditors compared to those of non-breach auditors. When switching auditors, clients are more likely to choose an auditor who has not experienced a data breach in recent years (typically within the past 3–5 years). We also find that a data breach is negatively associated with audit quality – breach-affected auditors become less conservative and allow greater discretion in financial reporting during the breach year. Furthermore, auditors increase audit fees following exposure to a data breach. Overall, our findings suggest that auditor-level cybersecurity breaches impair auditors’ reputations and impose significant costs.

Precise Authentication Watermarking Algorithm Based on Multiple Sorting Mechanisms for Vector Geographic Data

Symmetry-breaking in security mechanisms can create vulnerabilities which attackers may exploit to gain unauthorized access or cause data leakage, ultimately compromising the integrity and security of vector geographic data. How to achieve tamper localization remains a challenging task in the field of data authentication research. We propose a precise authentication watermarking algorithm for vector geographic data based on multiple sorting mechanisms. During the watermark embedding process, a sequence of points is initially extracted from the original data, followed by embedding watermarks into each coordinate point. The embedded watermark information consists of the self-identification and ordering information of each coordinate point. Ordering information is crucial for establishing relationships among points and enhancing tamper localization. During the authentication phase, the extracted watermark information is compared with the newly generated watermark information. Self-identification information is used to authenticate addition attacks, while ordering information is used to authenticate deletion attacks. Experimental results demonstrate that the proposed algorithm achieves high precision in detecting and localizing both addition and deletion attacks, significantly outperforming the comparison method.

Detecting Data Leaks due to SQL Injection

Abstract—SQL injection is a prevalent web application vulner- ability that allows attackers to manipulate database queries, often leading to severe data leaks. This paper surveys existing research on SQL injection attacks and various detection and prevention techniques. We explore classical and modern approaches, includ- ing machine learning paradigms, input verification methods, and parameterized query techniques. By examining the strengths and weaknesses of these approaches, this paper proposes a compre- hensive prevention framework. The challenges, limitations, and future directions for enhancing SQL injection security are also discussed. Index Terms—SQL, vulnerability, queries, leaks, approaches, prevention, machine learning, parameterized query, framework, injection

The Corporate Legal Responsibility for The Leak of Personal Data of Application Consumers in Indonesia

The importance of using identity in the form of personal data in all activities related to the fulfillment of rights encourages someone to provide personal data information. Voluntary and mandatory giving is a shame for individuals who seek more profit by using someone's personal data without the permission of the owner of the personal data concerned. The scope and formulation of this research problem focuses on the use of personal data without permission which can be qualified as a criminal offense and what criminal responsibility there is for perpetrators who use personal data without permission. The method used in this writing is the juridical-normative-conceptual legal research method. Research shows that the use of personal data without permission is a criminal offense by fulfilling the elements of general criminal regulations and related special regulations below. In Indonesia, criminal responsibility for perpetrators still does not have specific regulations regarding the imposition of criminal sanctions, so that responsibility is contained in separate regulations. The separate regulations contain several aspects of the protection of a person's personal data in general. The imposition of a crime against a corporation and/or its management does not preclude the possibility of a criminal being imposed on other perpetrators who, based on the provisions of the Law, are proven to be involved in the crime. Based on this research, the government, as the protector and guarantor of citizens' rights regarding the importance of personal data, must immediately ratify the Draft Law on Personal Data Protection for the sake of legal certainty.