Machine Learning, Deep Learning and AI for Cybersecurity

• Titel: Machine Learning, Deep Learning and AI for Cybersecurity
• Mitwirkende: Stamp, Mark [HerausgeberIn]
• Mitwirkende: Jureček, Martin [HerausgeberIn]
• Verf.angabe: edited by Mark Stamp, Martin Jureček
• Ausgabe: 1st ed. 2025.
• Verlagsort: Cham
• Verlagsort: Cham
• Verlag: Springer Nature Switzerland
• Verlag: Imprint: Springer
• E-Jahr: 2025
• Jahr: 2025.
• Jahr: 2025.
• Umfang: 1 Online-Ressource(IX, 647 p. 224 illus., 206 illus. in color.)
• ISBN: 978-3-031-83157-7
• DOI: doi:10.1007/978-3-031-83157-7
• Datenträger: Online-Ressource
• Sprache: eng
• Bibliogr. Hinweis: Erscheint auch als : Druck-Ausgabe
• Bibliogr. Hinweis: Erscheint auch als : Druck-Ausgabe
• Bibliogr. Hinweis: Erscheint auch als : Druck-Ausgabe
• K10plus-PPN: 1925484823
• K10plus-PPN:
  Universitätsbibliothek
• Lokale URL UB: Zum Volltext

Abstract

Online Clustering of Known and Emerging Malware Families -- Applying Word Embeddings and Graph Neural Networks for Effective Malware Classification -- A Comparative Analysis of SHAP and LIME in Detecting Malicious URLs -- Comparing Balancing Techniques for Malware Classification -- Multimodal Deception and Lie Detection Using Linguistic and Acoustic Features, Deep Models, and Large Language Models -- Enhancing Dynamic Keystroke Authentication with GAN-Optimized Deep Learning Classifiers -- Selecting Representative Samples from Malware Datasets -- FLChain: Integration of Federated Learning and Blockchain for Building Unified Models for Privacy Preservation -- On the Steganographic Capacity of Selected Learning Models -- An Empirical Analysis of Federated Learning Models Subject to Label-Flipping Adversarial Attack -- An Empirical Analysis of Hidden Markov Models with Momentum -- Image-Based Malware Classification Using QR and Aztec Codes -- Keystroke Dynamics for User Identification -- Distinguishing Chatbot from Human -- Malware Classification using a Hybrid Hidden Markov Model-Convolutional Neural Network -- Temporal Analysis of Adversarial Attacks in Federated Learning -- Steganographic Capacity of Transformer Models -- Robustness of Selected Learning Models under Label Flipping Attacks -- Effectiveness of Adversarial Benign and Malware Examples in Evasion and Poisoning Attacks -- Quantum Computing Methods for Malware Detection -- Reducing the Surface for Adversarial Attacks in Malware Detectors -- XAI and Android Malware Models.

Abstract

This book addresses a variety of problems that arise at the interface between AI techniques and challenging problems in cybersecurity. The book covers many of the issues that arise when applying AI and deep learning algorithms to inherently difficult problems in the security domain, such as malware detection and analysis, intrusion detection, spam detection, and various other subfields of cybersecurity. The book places particular attention on data driven approaches, where minimal expert domain knowledge is required. This book bridges some of the gaps that exist between deep learning/AI research and practical problems in cybersecurity. The proposed topics cover a wide range of deep learning and AI techniques, including novel frameworks and development tools enabling the audience to innovate with these cutting-edge research advancements in various security-related use cases. The book is timely since it is not common to find clearly elucidated research that applies the latest developments in AI to problems in cybersecurity.