Ehsan Nowroozi’s Post

The Technology Monitoring team at the Cyber-Defence Campus released a book dedicated to Large Language Models in Cybersecurity. The first book titled "Trends in Data Protection and Encryption Technologies" appeared in 2023. This book series provides advances in technology and trends for government, industry, and academic decision-makers, as well as technical experts. It echoes our extensive work on the Commission's projects, addressing public governance, safety and cyber resilience, digital and physical infrastructure oversight. In particular, it addresses: ➡ Overview of existing LLM families, adapting LLMs to downstream applications, conversational agents, tasks, evaluation and limitations of LLMs; ➡ LLMs in cybersecurity, including private information leakages, phishing and social engineering, flow hijacking, vulnerabilities, influence operations; ➡ Tracking and forecasting exposure, adoption trends and associated risks; ➡ Mitigation, including enchanting security awareness and education, adversarial evasion, robust and private federated learning, red teaming, security and privacy standards. Book - https://lnkd.in/en6V7YFd Authors and contributors: Andrei KUCHARAVY Valentin Mulder Vincent Lenders Octave Plancherel Alain Mermoud Florian Schuetz Jan Kleijssen Ljiljana Dolamic Natalia Ostapuk Julien Audiffren Beat Buesser Sebastiano Panichella Terry Vogelsang Zachary Schillaci, PhD #ai #ethics #policy #llm

Today's highlight is Shoshana Sugerman! If the RCC had a lead researcher position, it would definitely be Shoshana. She is always working on two or three different high-impact #research projects at the same time. Whether it's #quantum machine learning, exploring the ethics of Generative #AI on human-facing #cybersecurity tasks, and now evaluating compiler-based mitigation techniques, she is always excited to be contributing to the expansion of the state-of-the-art in cybersecurity knowledge. #security #cybersec #infosec #informationsecurity #generativeai #ml #machinelearning #RPI Rensselaer Polytechnic Institute Rensselaer Polytechnic Institute Information Technology & Web Science

🚀 New Publication! 🚀 With Mustafa Habeeb, A.S. Albahri (Ahmed Shihab Albahri) Open Access! "Network and Cybersecurity Applications of Defense in Adversarial Attacks: A State-of-the-Art Using Machine Learning and Deep Learning Methods" ------------------- This comprehensive study delves into the defense strategies against adversarial attacks using cutting-edge machine learning and deep learning techniques. By systematically reviewing 704 papers from reputable sources like ScienceDirect, Scopus, IEEE Xplore, and Web of Science, we identified key methodologies and emerging trends in the field. 🌟 Key Contributions: In-depth Investigation: Exploring adversarial attack generation, defensive strategies, and applications in malware, intrusion, and anomaly detection. Literature Analysis: Highlighting current research trends, challenges, and future directions in network and cybersecurity defense. Research Gaps: Identifying significant gaps and proposing a roadmap for enhancing defense mechanisms. 🔑 Implications: Our findings offer valuable insights for researchers, practitioners, and policymakers, paving the way for more robust and reliable defense strategies against advanced persistent threats. This work underscores the potential of AI and deep learning in revolutionizing cybersecurity. 📖 Read the full paper here: https://lnkd.in/dfTTr_Tz -------------- #Cybersecurity #MachineLearning #DeepLearning #Adversarial_Attacks #Research #Publication #AI #Network #Security #Innovation

We are thrilled to announce that our latest study, "Mitigating Label Flipping Attacks in Malicious URL Detectors Using Ensemble Trees," is now accessible on Arxiv [https://lnkd.in/dNdCqcFJ] and is currently undergoing submission to a journal for publication. A collaborative effort with Nada Jadalla, Samaneh Ghelichkhani, and Dr. Alireza Jolfaei. This study addresses the threat posed by malicious URLs across industries, such as transportation, healthcare, energy, and banking, emphasizing the importance of detecting these URLs to prevent business disruptions. However, current Machine Learning (ML) models are vulnerable to backdoor attacks such as Label Flipping (LF), which manipulates training data labels, leading to misclassification. To counter these attacks, this study proposes the integration of defense mechanisms into ML models. Specifically, it focuses on backdoor attacks in URL detection using ensemble trees and proposes an alarm system to detect poisoned labels. Experimental results using Alexa and Phishing Site URL datasets demonstrate the efficacy of the proposed defense mechanism in detecting and mitigating LF attacks. [If you have expertise in artificial intelligence for security and programming, I am keen to embark on new research collaborations. Do not hesitate to contact me via email at e.nowroozi@rave.ac.uk if our research resonates with you.] #DataPoisoning #ComputerNetworks #ResearchPublication #Collaboration #ArtificialIntelligence #Security #DataSecurity #ModelIntegrity #FLAttacks #LabelFlipping #FeaturePoisoning #AcademicResearch #AIforSecurity #Cybersecurity #InfoSec #MachineLearning #SecurityTech #ThreatDetection #SecurityAlgorithms #SecurityAutomation #SecurityAnalytics #AIinCybersecurity #SecurityResearch #SecuritySolutions #CyberDefense #SecurityInnovation #DataProtection #DigitalSecurity #AIandSecurity #SecureAI #SecurityFrameworks Ravensbourne University London

Excited to share my latest publication in the International Journal of Scientific Research in Engineering and Management (IJSREM)! 🎉 📄 Title: Phishing Website Detection Using Machine Learning In this study, we developed a method to detect phishing websites using machine learning. By analyzing URL features like length, special characters, subdomains, and keywords, we built a model that accurately identifies phishing URLs. This research highlights the methodology and the model's performance in real-world scenarios. Grateful for the support and guidance from my mentor, Ms. Gayathri N, and collaborators, Siva Sakthii U S and Janani Dhandayutham. 📖 Read the full paper here: https://lnkd.in/g-XMciH2 ISSN: 2582-3930 Impact Factor: 8.448 Volume 8 - Issue 07 July, 2024 #Research #Engineering #AI #MachineLearning #CyberSecurity #IJSREM #Publication #ScientificResearch

I'm excited to share that our paper, "STFL: Utilizing a Semi-Supervised, Transfer-Learning, Federated-Learning Approach to Detect Phishing URL Attacks" has been accepted for presentation at the International Joint Conference on Neural Networks (IJCNN) as part of IEEE WCCI, to be held at Pacifico Yokohama, Japan 🇯🇵, from June 30 to July 5, 2024. In this paper, we present a novel machine learning approach for phishing detection that combines semi-supervised learning, transfer learning, and federated learning. We train a Bi-LSTM autoencoder across decentralized edge devices containing unlabeled data. A centralized server collects the Bi-LSTM autoencoders and aggregates them into a global network using the FedAVG algorithm. The server then performs transfer learning to utilize the learned patterns from the global Bi-LSTM autoencoder networks and induce a classification model. Our experiments demonstrate that our proposed approach outperforms existing methods. I would like to thank all co-authors Prof. Yuval Elovici , Prof. Asaf Shabtai and Dr. Edita Grolman, for their help and guidance. Looking forward to presenting our paper at the conference. #IJCNN2024 #IEEEWCCI #DataScience #MachineLearning #CyberSecurity #FederatedLearning

I'm thrilled to share the exciting news of our recent publication in the prestigious International Journal of Scientific Research in Engineering and Management (IJSREM)! Our paper delves into the critical realm of cybersecurity with a focused survey on Machine Learning Based Intrusion Detection Systems. In today's rapidly evolving digital landscape, the need for robust intrusion detection mechanisms has never been more pressing. With cyber threats becoming increasingly sophisticated and pervasive, traditional rule-based approaches often fall short in detecting and preventing intrusions effectively. Hence, our research endeavors to explore the potential of machine learning algorithms in enhancing the efficiency and accuracy of intrusion detection systems. Through a meticulous review of the latest literature and methodologies, our survey offers valuable insights into the state-of-the-art techniques employed in the design and implementation of machine learning-based intrusion detection systems. We examine various machine learning algorithms, including but not limited to neural networks, support vector machines, decision trees, and ensemble methods, to discern their strengths, weaknesses, and suitability in different intrusion detection scenarios. Furthermore, we investigate the integration of advanced data preprocessing techniques, feature selection methods, and anomaly detection approaches to optimize the performance of these systems in detecting both known and unknown threats. By synthesizing the findings from diverse research studies and real-world applications, our paper aims to provide practitioners, researchers, and cybersecurity professionals with a comprehensive understanding of the evolving landscape of intrusion detection. We highlight the challenges and opportunities associated with the adoption of machine learning in cybersecurity and discuss potential avenues for future research and innovation. Ultimately, our endeavor is to contribute to the advancement of cybersecurity practices by equipping stakeholders with the knowledge and tools necessary to combat emerging cyber threats effectively. Dive into our research to explore the forefront of machine learning-based intrusion detection and join us in safeguarding digital environments against evolving security risks. for view : https://lnkd.in/d3BfyD_c for download : https://lnkd.in/dJFQDEpC #MachineLearning #Cybersecurity #Research 📝🔒

🔐 Strengthening Cybersecurity with Advanced Deep Learning Models 🔐 I’m excited to share our latest research where we introduce a hybrid DL-based model that combines Long Short-Term Memory (LSTM) and Residual Networks (ResNet) for intrusion detection systems (IDSs). In the face of increasing and more sophisticated cyber-attacks, the need for robust intrusion detection systems (IDSs) has never been more critical. An IDS serves as the first line of defense against online threats, playing a pivotal role in identifying and mitigating potential breaches. Our proposed model is designed to distinguish between normal activity and various types of attacks, including DoS, U2R, R2L, and probe attacks, using the NSL-KDD dataset. We’re thrilled to report that our model achieves a remarkable accuracy of 99.5%! Additionally, it outperforms other machine learning (ML) and DL models, setting a new benchmark in network intrusion detection. For those interested in the technical details, you can read our full paper here. https://lnkd.in/dHAVD-Ub Curious to see the code in action? Check out our GitHub repository here. https://lnkd.in/d4JYdCNS #CyberSecurity #DeepLearning #IntrusionDetection #MachineLearning #TechInnovation #AI #LSTM #ResNet #Research #NSLKDD #TechCommunity #GitHub

Excited to share my latest publication in the International Journal of Scientific Research in Engineering and Management (IJSREM)! 🎉 📄 Title: Phishing Website Detection Using Machine Learning In this study, we developed a method to detect phishing websites using machine learning. By analyzing URL features like length, special characters, subdomains, and keywords, we built a model that accurately identifies phishing URLs. This research highlights the methodology and the model's performance in real-world scenarios. Grateful for the support and guidance from my mentor, Ms. Gayathri N, and collaborators, Pradeepa Murugesan and Janani Dhandayutham. 📖 Read the full paper here: https://lnkd.in/g-XMciH2 ISSN: 2582-3930 Impact Factor: 8.448 Volume 8 - Issue 07 July, 2024 #Research #Engineering #AI #MachineLearning #CyberSecurity #IJSREM #Publication #ScientificResearch

New skills to deploy: detecting deepfake. Fake images or fake voices. Soon fake digital prints and biometrics?