Aug 2026· International Journal of Interactive Mobile Technologies (ijim)· Vol 20· 0 citations· 7 references
TL;DR
This proposed framework aims to fortify data protection and ensure user privacy in essential areas like healthcare, financial services, and e-governance, thereby fostering increased trust.
Abstract
The rapid growth of cyber landscapes and the development of a new cybersecurity model incorporating PET, deep learning, fuzzy systems, keystroke dynamic authentication, and encryption. It is used to prevent attacks by malware or unauthorized access to cloud systems. The proposed framework, which integrates an artificial intelligence (AI)-driven approach with identity and access management (IAM), enables the adaptive implementation of risk-based login authentication and real time anomaly detection. Unlike conventional security systems that depend on fixed rules and signatures, we provide more sophisticated solutions. A Floydel firewall is dynamically tailored through deep neural networks (DNNs) and automatically adjusts to fluctuating traffic patterns. It employs malware classification based on behavior, utilizes fuzzy logic to manage uncertainty during intrusions, and uses keystroke dynamics for user verification through typing patterns. The experiment demonstrates a 97.6% detection accuracy on benchmark data, while significantly reducing false positives and ensuring data confidentiality through encryption. The nature of cloud security can evolve based on the specific circumstances and threats we encounter. Looking ahead, we plan to delve into cryptography and distributed training to bolster decentralized infrastructures. This proposed framework aims to fortify data protection and ensure user privacy in essential areas like healthcare, financial services, and e-governance, thereby fostering increased trust.
An incremental learning based framework to detect anomalous traffic patterns which may indicate any key misuse in Software-Defined Networks in dynamic and real-time environments in modern SDN environments is proposed.
Gineeth Rajeshkhanna, Tamilarasi Kathirvel Murugan, Logeswari Govindaraj et al.· Journal of Computer Virology...· 0 citations
The exponential rise in cyber threats has created a critical need for intelligent and adaptive intrusion detection systems
(IDS) capable of identifying both known and emerging attack patterns. Traditional rule-based IDS mechanisms, such as Snort,
rely heavily on predefined signatures and struggle against sophisticated attacks including port scanning, web-based exploits, and
distributed denial-of-service (DDoS) attacks. This paper presents an AI-based Intrusion Detection System that integrates
network simulation, machine learning, and real-time visualization into a unified three-layer framework. The NS-3 network
simulator generates realistic normal and malicious traffic between attacker, router, and victim nodes; the resulting packetcapture (PCAP) data is processed by a Python-based IDS engine that applies signature rules for port scanning, DoS flooding,
and web attacks (SQL Injection, XSS, LFI, command injection); and a Random Forest classifier, trained on the CIC-IDS2017
benchmark dataset, augments detection with machine-learning-based classification. A Flask-based web dashboard provides realtime visualization of alerts, packet statistics, and attack distribution. Experimental results show an average detection accuracy of
98.5%, an average F1-score of 97.7%, and a false-positive rate below 1.2%, outperforming rule-based and prior deep-learning
baselines on comparable attack categories. The proposed multi-layered architecture demonstrates that combining simulation,
machine learning, and visualization can produce a scalable and effective solution for modern network security challenges.
T. Senthil, V. Shanmuganeethi· International Journal for Re...· 0 citations
A unified machine-learning framework for defence against malware-laden URLs, which simultaneously targets intrusion detection and intrusion prevention through module-aware, bio-inspired feature selection through module-aware, bio-inspired feature selection.
Mohammad Sh. Daoud, M. Abualhaj, Sumaya S. Al-Khatib et al.· Journal of Information and C...· 0 citations
This review presents a comprehensive analysis of machine learning-based intrusion detection systems, covering a wide range of techniques including supervised learning, unsupervised learning, ensemble learning, and deep learning models, and discusses critical challenges affecting the deployment of ML-based IDS.
Ranobir Hasan, H. Jamal, Kamal Kamal et al.· The Eastasouth Journal of In...· 0 citations
The paper examines the integration of the clever honeypots with the attacker behaviour analysis to enhance the network intrusion detection in the contemporary cyberspace environment. Due to the rapid development of cyber threats, the target of traditional intrusion detection systems (that are primarily based on fixed rules and known signatures) is to identify zero-day exploits, polymorphism, malware, and advanced persistent threats. In an attempt to circumvent these constraints, the study employs a quantitative paradigm that incorporates the survey of experts, simulated honeypot logs, and machine-learning-based behaviour analysis. Honeypot data around the world show trends in the frequency of attacks and ports attacked as well as the geographic origin and time. The 4 machine-learning models, Logistic Regression, KNN, Random Forest and XGBoost, were trained and tested on the feature-engineered datasets. The ensemble techniques performed well as compared to their linear counterparts with XGBoost recording 99.96 0-percent accuracy and Rand. Forest recording a 100 percent accuracy in the dataset. The findings demonstrate that intelligent honeypots do not only collect valuable behavioural indications, but also offer highly predictive attributes to automated detecting mechanisms. The research concludes that combining honeypot intelligence with machine learning improves real-time identification, proactive defence, as well as reducing the false alarms.
Unknown authors· Journal of Superintelligence...· 0 citations
Ransomware threats continue to evade traditional signature-based security strategies, particularly when
exploiting zero-day attacks, polymorphic methods, and code obfuscation. Rather than relying on static file analysis, the
system continuously manages runtime process behavior by analyzing key indicators of ransomware operation, including file
encryption rates, mass file renaming, entropy fluctuations, registry modifications, and network connections. This dynamic
behavioral analysis enables the timely identification of malicious activities, consequently enhancing the system's capability
to recognize ransomware threats in real time. The identification engine was implemented using React and TypeScript and
uses a configurable, weighted rule-based scoring strategy to classify running processes as either malicious. During simulated
assessments involving well-known ransomware families, including WannaCry, LockBit3, and Ryuk, the application
efficiently differentiated malicious processes from legitimate ones, delivering a identification accuracy of 88.9% while
maintaining a low false-positive rate. In addition, the proposed solution indicated real-time responsiveness, with an average
event update latency of approximately 360 milliseconds. The experimental results show that the behavior-based
identification methods generates more effective coverage against novel, polymorphic, and fileless ransomware threats than
conventional signature-based identification approaches. Based on these results, it is suggested that the behavioral
identification engine be combined into Endpoint Identification and Response (EDR) platforms to promote intelligent threat
containment, increase incident response, and reduce the danger of data loss.
Kazeem O. N., Abdul Kareem Olaitan Mummen, Shamsudeen Sani Saleh· International Journal of Inn...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
