Kim / Aminanto / Tanuwidjaja Network Intrusion Detection using Deep Learning

A Feature Learning Approach
1. Auflage 2018
ISBN: 978-981-13-1444-5
Verlag: Springer Nature Singapore
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 79 Seiten

Reihe: Computer Science (R0)

This book presents recent advances in intrusion detection systems (IDSs) using state-of-the-art deep learning methods. It also provides a systematic overview of classical machine learning and the latest developments in deep learning.  In particular, it discusses deep learning applications in IDSs in different classes: generative, discriminative, and adversarial networks. Moreover, it compares various deep learning-based IDSs based on benchmarking datasets. The book also proposes two novel feature learning models: deep feature extraction and selection (D-FES) and fully unsupervised IDS. Further challenges and research directions are presented at the end of the book.

Offering a comprehensive overview of deep learning-based IDS, the book is a valuable reerence resource for undergraduate and graduate students, as well as researchers and practitioners interested in deep learning and intrusion detection. Further, the comparison of various deep-learning applications helps readers gain a basic understanding of machine learning, and inspires applications in IDS and other related areas in cybersecurity.

Autoren/Hrsg.

Kim, Kwangjo

Aminanto, Muhamad Erza

Tanuwidjaja, Harry Chandra

Inhaltsverzeichnis

1;Preface;7
2;Acknowledgments;9
3;Contents;10
4;Acronyms;13
5;1 Introduction;16
5.1;References;19
6;2 Intrusion Detection Systems;20
6.1;2.1 Definition;20
6.2;2.2 Classification;20
6.3;2.3 Benchmark;23
6.3.1;2.3.1 Performance Metric;23
6.3.2;2.3.2 Public Dataset;24
6.4;References;25
7;3 Classical Machine Learning and Its Applications to IDS;27
7.1;3.1 Classification of Machine Learning;27
7.1.1;3.1.1 Supervised Learning;27
7.1.1.1;3.1.1.1 Support Vector Machine;27
7.1.1.2;3.1.1.2 Decision Tree;28
7.1.2;3.1.2 Unsupervised Learning;29
7.1.2.1;3.1.2.1 K-Means Clustering;29
7.1.2.2;3.1.2.2 Ant Clustering;29
7.1.2.3;3.1.2.3 (Sparse) Auto-Encoder;30
7.1.3;3.1.3 Semi-supervised Learning;33
7.1.4;3.1.4 Weakly Supervised Learning;34
7.1.5;3.1.5 Reinforcement Learning;34
7.1.6;3.1.6 Adversarial Machine Learning;35
7.2;3.2 Machine-Learning-Based Intrusion Detection Systems;35
7.3;References;38
8;4 Deep Learning;41
8.1;4.1 Classification;41
8.2;4.2 Generative (Unsupervised Learning);41
8.2.1;4.2.1 Stacked (Sparse) Auto-Encoder;42
8.2.2;4.2.2 Boltzmann Machine;44
8.2.3;4.2.3 Sum-Product Networks;44
8.2.4;4.2.4 Recurrent Neural Networks;44
8.3;4.3 Discriminative;46
8.4;4.4 Hybrid;46
8.4.1;4.4.1 Generative Adversarial Networks (GAN);46
8.5;References;47
9;5 Deep Learning-Based IDSs;49
9.1;5.1 Generative;49
9.1.1;5.1.1 Deep Neural Network;49
9.1.2;5.1.2 Accelerated Deep Neural Network;50
9.1.3;5.1.3 Self-Taught Learning;51
9.1.4;5.1.4 Stacked Denoising Auto-Encoder;52
9.1.5;5.1.5 Long Short-Term Memory Recurrent Neural Network;52
9.2;5.2 Discriminative;53
9.2.1;5.2.1 Deep Neural Network in Software-Defined Networks;53
9.2.2;5.2.2 Recurrent Neural Network;54
9.2.3;5.2.3 Convolutional Neural Network;54
9.2.4;5.2.4 Long Short-Term Memory Recurrent Neural Network;55
9.2.4.1;5.2.4.1 LSTM-RNN Staudemeyer;55
9.2.4.2;5.2.4.2 LSTM-RNN for Collective Anomaly Detection;55
9.2.4.3;5.2.4.3 GRU in IoT;55
9.2.4.4;5.2.4.4 LSTM-RNN for DDoS;56
9.3;5.3 Hybrid;56
9.3.1;5.3.1 Adversarial Networks;56
9.4;5.4 Deep Reinforcement Learning;57
9.5;5.5 Comparison;57
9.6;References;58
10;6 Deep Feature Learning;60
10.1;6.1 Deep Feature Extraction and Selection;60
10.1.1;6.1.1 Methodology;61
10.1.2;6.1.2 Evaluation;65
10.1.2.1;6.1.2.1 Dataset Preprocessing;65
10.1.2.2;6.1.2.2 Experimental Result;66
10.2;6.2 Deep Learning for Clustering;72
10.2.1;6.2.1 Methodology;75
10.2.2;6.2.2 Evaluation;76
10.3;6.3 Comparison;78
10.4;References;80
11;7 Summary and Further Challenges;82
11.1;References;83
12;Appendix A A Survey on Malware Detection from Deep Learning;84
12.1;A.1 Automatic Analysis of Malware BehaviorUsing Machine Learning;84
12.2;A.2 Deep Learning for Classification of Malware System Call Sequences;85
12.3;A.3 Malware Detection with Deep Neural Network Using Process Behavior;86
12.4;A.4 Efficient Dynamic Malware Analysis Based on Network Behavior Using Deep Learning;86
12.5;A.5 Automatic Malware Classification and New Malware Detection Using Machine Learning;87
12.6;A.6 DeepSign: Deep Learning for Automatic Malware Signature Generation and Classification;88
12.7;A.7 Selecting Features to Classify Malware;88
12.8;A.8 Analysis of Machine-Learning Techniques Used in Behavior-Based Malware Detection;89
12.9;A.9 Malware Detection Using Machine-Learning-Based Analysis of Virtual Memory Access Patterns;90
12.10;A.10 Zero-Day Malware Detection;90
12.11;References;91