Intrusion Detection: A Data Mining Approach

Description

This book presents state-of-the-art research on intrusion detection using reinforcement learning, fuzzy and rough set theories, and genetic algorithm. Reinforcement learning is employed to incrementally learn the computer network behavior, while rough and fuzzy sets are utilized to handle the uncertainty involved in the detection of traffic anomaly to secure data resources from possible attack. Genetic algorithms make it possible to optimally select the network traffic parameters to reduce the risk of network intrusion. The book is unique in terms of its content, organization, and writing style. Primarily intended for graduate electrical and computer engineering students, it is also useful for doctoral students pursuing research in intrusion detection and practitioners interested in network security and administration. The book covers a wide range of applications, from general computer security to server, network, and cloud security. Preface......Page 6 List of Publication by Dr. Nandita Sengupta Relevant to the Book......Page 8 List of Publication by Prof. Jaya Sil Relevant to the Book......Page 10 Acknowledgements......Page 14 Contents......Page 15 About the Authors......Page 18 1 Introduction......Page 20 1.1.1 Types of IDS......Page 22 1.1.2 Types of Attacks......Page 24 1.2.1 Cleaning of Data......Page 25 1.2.4 Data Reduction......Page 27 1.3.1 Classification of Discretization Methods......Page 30 1.3.2 Methods of Discretization......Page 31 1.4.1 Dynamic Learning......Page 33 1.5 The Work......Page 34 1.5.1 Contributions......Page 35 1.6 Summary......Page 37 References......Page 38 2.1 Preprocessing......Page 45 2.2 Cut Generation Method......Page 46 2.2.1 Algorithm for Generation of Cut......Page 47 2.2.3 Discrete Value Mapping......Page 50 2.3.1 Optimized Equal Width Interval (OEWI)......Page 52 2.3.2 Split and Merge Interval (SMI)......Page 55 2.4 Discussions on Results......Page 57 2.5 Summary......Page 58 References......Page 62 3 Data Reduction......Page 65 3.1.1 Preliminaries of RST......Page 66 3.1.2 Reduct Using Discernibility Matrix......Page 69 3.1.3 Reduct Using Attribute Dependency......Page 73 3.2.1 Fuzzy–Rough Sets......Page 76 3.2.2 Rule-Base......Page 78 3.2.3 Fuzzy–Rough–GA......Page 81 3.3 Instance Reduction......Page 85 3.3.1 Simulated Annealing-Based Clustering Algorithm......Page 86 3.3.2 ModifiedSAFC Algorithm......Page 87 3.3.3 Most Significant Cluster......Page 89 3.4.1 Results of Dimension Reduction on Discrete Domain......Page 90 3.4.2 Confusion Matrix......Page 93 3.4.3 Results of Dimension Reduction on Continuous Domain......Page 94 3.4.4 Accuracy After Instance Reduction......Page 95 References......Page 97 4.1 Q-Learning......Page 101 4.1.1 Extended-Q-Learning Algorithm for Optimized Cut Generation......Page 103 4.2 Hierarchical-Q-Learning Approach......Page 113 4.2.2 Optimization of Linguistic Labels......Page 114 4.3.1 Result of Extended-Q-Learning Algorithm......Page 116 4.3.2 Experiments Using Synthetic Data Set......Page 119 4.3.3 Results of the Proposed Hierarchical-Q-Learning Algorithm......Page 122 4.4 Summary......Page 124 References......Page 127 5.1 Essence of the Proposed Methods......Page 130 5.2 Outstanding Issues......Page 131 5.3 Future Research Directions......Page 133 References......Page 134 Network Traffic Data Set......Page 136 References......Page 145 Subject Index......Page 146