Machine Learning in Cyber Trust

by Tsai, Jeffrey J. P.; Yu, Philip S.

• ISBN13: 9780387887340
• ISBN10: 0387887342
• Format: Hardcover
• Copyright: 2009-04-17
• Publisher: Springer-Verlag New York Inc

Summary

Many networked computer systems are far too vulnerable to cyber attacks that can inhibit their functioning, corrupt important data, or expose private information. Not surprisingly, the field of cyber-based systems is a fertile ground where many tasks can be formulated as learning problems and approached in terms of machine learning algorithms.This book contains original materials by leading researchers in the area and covers applications of different machine learning methods in the reliability, security, performance, and privacy issues of cyber space. It enables readers to discover what types of learning methods are at their disposal, summarizing the state-of-the-practice in this significant area, and giving a classification of existing work.Those working in the field of cyber-based systems, including industrial managers, researchers, engineers, and graduate and senior undergraduate students will find this an indispensable guide in creating systems resistant to and tolerant of cyber attacks.

Table of Contents

Preface	p. vii
Cyber System	p. 1
Cyber-Physical Systems: A New Frontier	p. 3
Introduction	p. 3
The Challenges of Cyber-Physical System Research	p. 6
Real-time System Abstractions	p. 7
Robustness of Cyber-Physical Systems	p. 8
System QoS Composition Challenge	p. 10
Knowledge Engineering in Cyber-Physical Systems	p. 10
Medical Device Network: An Example Cyber-Physical System	p. 11
Summary	p. 12
References	p. 13
Security	p. 15
Misleading Learners: Co-opting Your Spam Filter	p. 17
Introduction	p. 17
Background	p. 19
Training Model	p. 20
The Contamination Assumption	p. 21
SpamBayes Learning Method	p. 21
Attacks	p. 23
Causative Availability Attacks	p. 25
Causative Integrity Attacks-Pseudospam	p. 28
Experiments	p. 29
Experimental Method	p. 29
Dictionary Attack Results	p. 33
Focused Attack Results	p. 36
Pseudospam Attack Experiments	p. 40
A Defense: Reject on Negative Impact (RONI)	p. 43
Related Work	p. 45
Conclusion	p. 46
Appendix: Analysis of an Optimal Attack	p. 46
Properties of the Spam Score	p. 47
Effect of Poisoning on Token Scores	p. 48
References	p. 50
Survey of Machine Learning Methods for Database Security	p. 53
Introduction	p. 53
Paper Road Map	p. 56
Detection of SQL Injection Attacks	p. 56
A Learning-based Approach to the Detection of SQL Attacks	p. 57
Profiling Database Applications to Detect SQL Injection Attacks	p. 59
Anomaly Detection for Defending Against Insider Threats	p. 60
DEMIDS: A Misuse Detection System for Database Systems	p. 60
A Data Mining Approach for Database Intrusion Detection	p. 61
Detecting Anomalous Database Access Patterns in Relational Databases	p. 62
Emerging Trends	p. 67
Database Activity Monitoring	p. 67
Responding to Database Anomalies	p. 68
Conclusion	p. 69
References	p. 70
Identifying Threats Using Graph-based Anomaly Detection	p. 73
Introduction	p. 73
Graph-based Learning	p. 75
Graph-based Anomaly Detection	p. 76
GBAD Approach	p. 78
Experimental Results	p. 81
Synthetic Data	p. 82
Real-world Datasets	p. 87
Other Domains	p. 103
Related Work	p. 104
Conclusions	p. 106
References	p. 107
On the Performance of Online Learning Methods for Detecting Malicious Executables	p. 109
Introduction	p. 109
Preliminaries	p. 111
Machine Learning	p. 111
Evaluation	p. 113
Detecting Malicious Executables	p. 114
Online Learning Methods	p. 115
Naive Bayes	p. 116
Stagger	p. 116
Winnow	p. 117
Hoeffding Tree	p. 118
Streaming Ensemble Algorithm	p. 118
Accuracy Weighted Ensemble	p. 119
Dynamic Weighted Majority	p. 119
From Executables to Examples	p. 120
Experimental Study	p. 121
Design	p. 121
Results and Analysis	p. 122
Discussion	p. 128
Concluding Remarks	p. 129
References	p. 130
Efficient Mining and Detection of Sequential Intrusion Patterns for Network Intrusion Detection Systems	p. 133
Introduction	p. 133
Existing Work	p. 136
The Proposed Network Intrusion Detection Framework	p. 137
C-RSPM Supervised Classification	p. 139
Temporal Pattern Analysis using LDM	p. 140
Experimental Setup	p. 145
Experimental Results	p. 147
Performance of C-RSPM	p. 147
Performance of LDM	p. 148
Performance of the Proposed Framework	p. 151
Conclusion	p. 152
Reference	p. 152
A Non-Intrusive Approach to Enhance Legacy Embedded Control Systems with Cyber Protection Features	p. 155
Introduction	p. 156
Related work	p. 159
Event-Based Non-intrusive Approach for Enhancing Legacy Systems with Self-Protection Features	p. 160
Motivating Example	p. 161
Control-loop architecture	p. 162
Observation Module	p. 164
Evaluation Module	p. 165
Protection Module	p. 166
Observation and Inference	p. 167
Making Decisions in Real-Time	p. 170
Truthful Voters	p. 172
Untruthful Voters	p. 174
Current Implementation	p. 177
Event and channels	p. 177
Modules	p. 177
Conclusion	p. 179
References	p. 180
Image Encryption and Chaotic Cellular Neural Network	p. 183
Introduction to image encryption	p. 183
Based on image scrambling technique	p. 184
Based on SCAN pattern	p. 185
Based on tree data structures	p. 185
Based on chaotic systems	p. 186
Image encryption scheme based on chaotic CNN	p. 188
Introduction to Cellular Neural Network	p. 188
Description of image encryption algoritham	p. 195
Security analyses	p. 199
Key space analysis	p. 199
Sensitivity analysis	p. 199
Information entropy	p. 202
Statistical analysis	p. 203
Comparisons with other chaos-based algorithms	p. 208
Conclusions and discussion	p. 209
References	p. 210
Privacy	p. 215
From Data Privacy to Location Privacy	p. 217
Introduction	p. 217
Data Privacy	p. 219
Models and Principles	p. 220
Techniques	p. 225
Location Privacy	p. 231
Models and Principles	p. 232
Location Anonymization Techniques	p. 237
Open Issues and Challenges	p. 241
Summary	p. 243
References	p. 244
Privacy Preserving Nearest Neighbor Search	p. 247
Introduction	p. 247
Overview	p. 249
Problem Description	p. 249
Definitions	p. 250
Secure Multiparty Computation Primitives	p. 251
Provable Security	p. 253
Nearest Neighbor Algorithm	p. 253
Nearest Neighbor Search	p. 253
Find Furthest Points	p. 258
Extension to the Multiparty Case	p. 261
Applications	p. 261
LOF Outlier Detection	p. 261
SNN Clustering	p. 264
kNN Classification	p. 265
Complexity Analysis	p. 268
Optimizations	p. 270
Related Work	p. 271
Conclusion	p. 273
References	p. 274
Reliability	p. 277
High-Confidence Compositional Reliability Assessment of SOA-Based Systems Using Machine Learning Techniques	p. 279
Introduction	p. 279
Related Work	p. 282
Reliability assessment of Service from Service Layer Perspective	p. 283
Reliability assessment of Composite Service from Service Layer Perspective	p. 284
Reliability assessment of Services from Component Layer Perspective	p. 285
Service Reliability Assessment Framework	p. 287
Component Reliability Assessment Framework	p. 288
System Reliability	p. 302
Experimental Studies	p. 311
Features of the ECM Application	p. 311
Dynamic Reliability Monitoring	p. 312
Storing the Testing and Dynamically Monitored Data	p. 313
Dynamic Monitoring for a Specific Service Invocation	p. 313
Conclusions	p. 319
References	p. 320
Model, Properties, and Applications of Context-Aware Web Services	p. 323
Introduction	p. 323
Related research	p. 326
Web Services and semantic Web	p. 326
Context and context-aware applications	p. 328
Context model	p. 329
Context description	p. 330
Context acquisition	p. 333
System implementation	p. 336
Context-aware applications	p. 339
Context-aware services discovery for finding right services	p. 339
Context-aware social collaborators discovery for finding right partners	p. 344
Context-aware content adaptation for finding right content presentation	p. 350
Conclusions	p. 355
References	p. 356
Index	p. 359
Table of Contents provided by Ingram. All Rights Reserved.