Artificial Intelligence and Data Mining Approaches in Security Frameworks
Description
Written and edited by a team of experts in the field, this outstanding new volume offers solutions to the problems of security, outlining the concepts behind allowing computers to learn from experience and understand the world in terms of a hierarchy of concepts, with each concept defined through its relation to simpler concepts.
Artificial intelligence (AI) and data mining is the fastest growing field in computer science. AI and data mining algorithms and techniques are found to be useful in different areas like pattern recognition, automatic threat detection, automatic problem solving, visual recognition, fraud detection, detecting developmental delay in children, and many other applications. However, applying AI and data mining techniques or algorithms successfully in these areas needs a concerted effort, fostering integrative research between experts ranging from diverse disciplines from data science to artificial intelligence. Successful application of security frameworks to enable meaningful, cost effective, personalized security service is a primary aim of engineers and researchers today. However realizing this goal requires effective understanding, application and amalgamation of AI and data mining and several other computing technologies to deploy such a system in an effective manner.
This book provides state of the art approaches of artificial intelligence and data mining in these areas. It includes areas of detection, prediction, as well as future framework identification, development, building service systems and analytical aspects. In all these topics, applications of AI and data mining, such as artificial neural networks, fuzzy logic, genetic algorithm and hybrid mechanisms, are explained and explored. This book is aimed at the modeling and performance prediction of efficient security framework systems, bringing to light a new dimension in the theory and practice.
This groundbreaking new volume presents these topics and trends, bridging the research gap on AI and data mining to enable wide-scale implementation. Whether for the veteran engineer or the student, this is a must-have for any library.
This groundbreaking new volume:
* Clarifies the understanding of certain key mechanisms of technology helpful in the use of artificial intelligence and data mining in security frameworks
* Covers practical approaches to the problems engineers face in working in this field, focusing on the applications used every day
* Contains numerous examples, offering critical solutions to engineers and scientists
* Presents these new applications of AI and data mining that are of prime importance to human civilization as a whole
More details
Other editions
Additional editions
Persons
Neeraj Bhargava, PhD, is a professor and head of the Department of Computer Science at Maharshi Dayanand Saraswati University in Ajmer, India, having earned his doctorate from the University of Rajasthan, Jaipur in India. He has over 30 years of teaching experience at the university level and has contributed to numerous books throughout his career. He has also published over 100 papers in scientific and technical journals and has been an organizing chair on over 15 scientific conferences. His work on face recognition and fingerprint recognition is often cited in other research and is well-known all over the world.
Ritu Bhargava, PhD, is an assistant professor in the Department of Computer Science at Sophia Girls College in Ajmer, India, having earned her PhD in computer science from Hemchandracharya North Gujarat University Patan, Gujarat, India. She has more than 15 years of active teaching and research experience and has contributed to three books and more than 30 papers in scientific and technical journals. She has also been an organizing chair on over 15 scientific conferences, and, like her colleague, her work on face recognition and fingerprint recognition is well-known and often cited.
Pramod Singh Rathore, MTech, is an assistant professor at the Aryabhatta College of Engineering and Research Center and visiting faculty member at MDSU in Ajmer, India. He is a PhD in computer science and engineering at the University of Engineering and Management and already has eight years of teaching experience and over 45 papers in scientific and technical journals. He has also co-authored and edited numerous books.
Rashmi Agrawal, PhD, is a professor in the Department of Computer Applications at the Manav Rachna International Institude of Research and Studies in Faridabad, India with more than 18 years of teaching experience. She is a book series editor and the associate editor on a scientific journal on data science and the internet of things. She has published many research papers in scientific and technical journals in these areas and contributed multiple chapters to numerous books. She is currently guiding PhD students and is an active reviewer and editorial board member of various journals.
Content
Preface xiii
1 Role of AI in Cyber Security 1
Navani Siroya and Prof Manju Mandot
1.1 Introduction 2
1.2 Need for Artificial Intelligence 2
1.3 Artificial Intelligence in Cyber Security 3
1.3.1 Multi-Layered Security System Design 3
1.3.2 Traditional Security Approach and AI 4
1.4 Related Work 5
1.4.1 Literature Review 5
1.4.2 Corollary 6
1.5 Proposed Work 6
1.5.1 System Architecture 7
1.5.2 Future Scope 7
1.6 Conclusion 7
References 8
2 Privacy Preserving Using Data Mining 11
Chitra Jalota and Dr. Rashmi Agrawal
2.1 Introduction 11
2.2 Data Mining Techniques and Their Role in Classification and Detection 14
2.3 Clustering 19
2.4 Privacy Preserving Data Mining (PPDM) 21
2.5 Intrusion Detection Systems (IDS) 22
2.5.1 Types of IDS 23
2.5.1.1 Network-Based IDS 23
2.5.1.2 Host-Based IDS 24
2.5.1.3 Hybrid IDS 25
2.6 Phishing Website Classification 26
2.7 Attacks by Mitigating Code Injection 27
2.7.1 Code Injection and Its Categories 27
2.8 Conclusion 28
References 29
3 Role of Artificial Intelligence in Cyber Security and Security Framework 33
Shweta Sharma
3.1 Introduction 34
3.2 AI for Cyber Security 36
3.3 Uses of Artificial Intelligence in Cyber Security 38
3.4 The Role of AI in Cyber Security 40
3.4.1 Simulated Intelligence Can Distinguish Digital Assaults 41
3.4.2 Computer-Based Intelligence Can Forestall Digital Assaults 42
3.4.3 Artificial Intelligence and Huge Scope Cyber Security 42
3.4.4 Challenges and Promises of Artificial Intelligence in Cyber Security 43
3.4.5 Present-Day Cyber Security and its Future with Simulated Intelligence 44
3.4.6 Improved Cyber Security with Computer-Based Intelligence and AI (ML) 45
3.4.7 AI Adopters Moving to Make a Move 45
3.5 AI Impacts on Cyber Security 46
3.6 The Positive Uses of AI Based for Cyber Security 48
3.7 Drawbacks and Restrictions of Using Computerized Reasoning For Digital Security 49
3.8 Solutions to Artificial Intelligence Confinements 50
3.9 Security Threats of Artificial Intelligence 51
3.10 Expanding Cyber Security Threats with Artificial Consciousness 52
3.11 Artificial Intelligence in Cybersecurity - Current Use-Cases and Capabilities 55
3.11.1 AI for System Danger Distinguishing Proof 56
3.11.2 The Common Fit for Artificial Consciousness in Cyber Security 56
3.11.3 Artificial Intelligence for System Danger ID 57
3.11.4 Artificial Intelligence Email Observing 58
3.11.5 Simulated Intelligence for Battling Artificial Intelligence Dangers 58
3.11.6 The Fate of Computer-Based Intelligence in Cyber Security 59
3.12 How to Improve Cyber Security for Artificial Intelligence 60
3.13 Conclusion 61
References 62
4 Botnet Detection Using Artificial Intelligence 65
Astha Parihar and Prof. Neeraj Bhargava
4.1 Introduction to Botnet 66
4.2 Botnet Detection 67
4.2.1 Host-Centred Detection (HCD) 68
4.2.2 Honey Nets-Based Detection (HNBD) 69
4.2.3 Network-Based Detection (NBD) 69
4.3 Botnet Architecture 69
4.3.1 Federal Model 70
4.3.1.1 IBN-Based Protocol 71
4.3.1.2 HTTP-Based Botnets 71
4.3.2 Devolved Model 71
4.3.3 Cross Model 72
4.4 Detection of Botnet 73
4.4.1 Perspective of Botnet Detection 73
4.4.2 Detection (Disclosure) Technique 73
4.4.3 Region of Tracing 74
4.5 Machine Learning 74
4.5.1 Machine Learning Characteristics 74
4.6 A Machine Learning Approach of Botnet Detection 75
4.7 Methods of Machine Learning Used in Botnet Exposure 76
4.7.1 Supervised (Administrated) Learning 76
4.7.1.1 Appearance of Supervised Learning 77
4.7.2 Unsupervised Learning 78
4.7.2.1 Role of Unsupervised Learning 79
4.8 Problems with Existing Botnet Detection Systems 80
4.9 Extensive Botnet Detection System (EBDS) 81
4.10 Conclusion 83
References 84
5 Spam Filtering Using AI 87
Yojna Khandelwal and Dr. Ritu Bhargava
5.1 Introduction 87
5.1.1 What is SPAM? 87
5.1.2 Purpose of Spamming 88
5.1.3 Spam Filters Inputs and Outputs 88
5.2 Content-Based Spam Filtering Techniques 89
5.2.1 Previous Likeness-Based Filters 89
5.2.2 Case-Based Reasoning Filters 89
5.2.3 Ontology-Based E-Mail Filters 90
5.2.4 Machine-Learning Models 90
5.2.4.1 Supervised Learning 90
5.2.4.2 Unsupervised Learning 90
5.2.4.3 Reinforcement Learning 91
5.3 Machine Learning-Based Filtering 91
5.3.1 Linear Classifiers 91
5.3.2 Naïve Bayes Filtering 92
5.3.3 Support Vector Machines 94
5.3.4 Neural Networks and Fuzzy Logics-Based Filtering 94
5.4 Performance Analysis 97
5.5 Conclusion 97
References 98
6 Artificial Intelligence in the Cyber Security Environment 101
Jaya Jain
6.1 Introduction 102
6.2 Digital Protection and Security Correspondences Arrangements 104
6.2.1 Operation Safety and Event Response 105
6.2.2 AI2 105
6.2.2.1 CylanceProtect 105
6.3 Black Tracking 106
6.3.1 Web Security 107
6.3.1.1 Amazon Macie 108
6.4 Spark Cognition Deep Military 110
6.5 The Process of Detecting Threats 111
6.6 Vectra Cognito Networks 112
6.7 Conclusion 115
References 115
7 Privacy in Multi-Tenancy Frameworks Using AI 119
Shweta Solanki
7.1 Introduction 119
7.2 Framework of Multi-Tenancy 120
7.3 Privacy and Security in Multi-Tenant Base System Using AI 122
7.4 Related Work 125
7.5 Conclusion 125
References 126
8 Biometric Facial Detection and Recognition Based on ILPB and SVM 129
Shubhi Srivastava, Ankit Kumar and Shiv Prakash
8.1 Introduction 129
8.1.1 Biometric 131
8.1.2 Categories of Biometric 131
8.1.2.1 Advantages of Biometric 132
8.1.3 Significance and Scope 132
8.1.4 Biometric Face Recognition 132
8.1.5 Related Work 136
8.1.6 Main Contribution 136
8.1.7 Novelty Discussion 137
8.2 The Proposed Methodolgy 139
8.2.1 Face Detection Using Haar Algorithm 139
8.2.2 Feature Extraction Using ILBP 141
8.2.3 Dataset 143
8.2.4 Classification Using SVM 143
8.3 Experimental Results 145
8.3.1 Face Detection 146
8.3.2 Feature Extraction 146
8.3.3 Recognize Face Image 147
8.4 Conclusion 151
References 152
9 Intelligent Robot for Automatic Detection of Defects in Pre-Stressed Multi-Strand Wires and Medical Gas Pipe Line System Using ANN and IoT 155
S K Rajesh Kanna, O. Pandithurai, N. Anand, P. Sethuramalingam and Abdul Munaf
9.1 Introduction 156
9.2 Inspection System for Defect Detection 158
9.3 Defect Recognition Methodology 162
9.4 Health Care MGPS Inspection 165
9.5 Conclusion 168
References 169
10 Fuzzy Approach for Designing Security Framework 173
Kapil Chauhan
10.1 Introduction 173
10.2 Fuzzy Set 177
10.3 Planning for a Rule-Based Expert System for Cyber Security 185
10.3.1 Level 1: Defining Cyber Security Expert System Variables 185
10.3.2 Level 2: Information Gathering for Cyber Terrorism 185
10.3.3 Level 3: System Design 186
10.3.4 Level 4: Rule-Based Model 187
10.4 Digital Security 188
10.4.1 Cyber-Threats 188
10.4.2 Cyber Fault 188
10.4.3 Different Types of Security Services 189
10.5 Improvement of Cyber Security System (Advance) 190
10.5.1 Structure 190
10.5.2 Cyber Terrorism for Information/Data Collection 191
10.6 Conclusions 191
References 192
11 Threat Analysis Using Data Mining Technique 197
Riddhi Panchal and Binod Kumar
11.1 Introduction 198
11.2 Related Work 199
11.3 Data Mining Methods in Favor of Cyber-Attack Detection 201
11.4 Process of Cyber-Attack Detection Based on Data Mining 204
11.5 Conclusion 205
References 205
12 Intrusion Detection Using Data Mining 209
Astha Parihar and Pramod Singh Rathore
12.1 Introduction 209
12.2 Essential Concept 210
12.2.1 Intrusion Detection System 211
12.2.2 Categorization of IDS 212
12.2.2.1 Web Intrusion Detection System (WIDS) 213
12.2.2.2 Host Intrusion Detection System (HIDS) 214
12.2.2.3 Custom-Based Intrusion Detection System (CIDS) 215
12.2.2.4 Application Protocol-Based Intrusion Detection System (APIDS) 215
12.2.2.5 Hybrid Intrusion Detection System 216
12.3 Detection Program 216
12.3.1 Misuse Detection 217
12.3.1.1 Expert System 217
12.3.1.2 Stamp Analysis 218
12.3.1.3 Data Mining 220
12.4 Decision Tree 221
12.4.1 Classification and Regression Tree (CART) 222
12.4.2 Iterative Dichotomise 3 (ID3) 222
12.4.3 C 4.5 223
12.5 Data Mining Model for Detecting the Attacks 223
12.5.1 Framework of the Technique 224
12.6 Conclusion 226
References 226
13 A Maize Crop Yield Optimization and Healthcare Monitoring Framework Using Firefly Algorithm through IoT 229
S K Rajesh Kanna, V. Nagaraju, D. Jayashree, Abdul Munaf and M. Ashok
13.1 Introduction 230
13.2 Literature Survey 231
13.3 Experimental Framework 232
13.4 Healthcare Monitoring 237
13.5 Results and Discussion 240
13.6 Conclusion 242
References 243
14 Vision-Based Gesture Recognition: A Critical Review 247
Neela Harish, Praveen, Prasanth, Aparna and Athaf
14.1 Introduction 247
14.2 Issues in Vision-Based Gesture Recognition 248
14.2.1 Based on Gestures 249
14.2.2 Based on Performance 249
14.2.3 Based on Background 249
14.3 Step-by-Step Process in Vision-Based 249
14.3.1 Sensing 251
14.3.2 Preprocessing 252
14.3.3 Feature Extraction 252
14.4 Classification 253
14.5 Literature Review 254
14.6 Conclusion 258
References 258
15 SPAM Filtering Using Artificial Intelligence 261
Abha Jain
15.1 Introduction 261
15.2 Architecture of Email Servers and Email Processing Stages 265
15.2.1 Architecture - Email Spam Filtering 265
15.2.1.1 Spam Filter - Gmail 266
15.2.1.2 Mail Filter Spam - Yahoo 266
15.2.1.3 Email Spam Filter - Outlook 267
15.2.2 Email Spam Filtering - Process 267
15.2.2.1 Pre-Handling 268
15.2.2.2 Taxation 268
15.2.2.3 Election of Features 268
15.2.3 Freely Available Email Spam Collection 269
15.3 Execution Evaluation Measures 269
15.4 Classification - Machine Learning Technique for Email Spam 275
15.4.1 Flock Technique - Clustering 275
15.4.2 Naïve Bayes Classifier 276
15.4.3 Neural Network 279
15.4.4 Firefly Algorithm 282
15.4.5 Fuzzy Set Classifiers 283
15.4.6 Support Vector Machine 284
15.4.7 Decision Tree 286
15.4.7.1 NBTree Classifier 286
15.4.7.2 C4.5/J48 Decision Tree Algorithm 287
15.4.7.3 Logistic Version Tree Induction (LVT) 287
15.4.8 Ensemble Classifiers 288
15.4.9 Random Forests (RF) 289
15.5 Conclusion 290
References 290
Index 295
Preface
Artificial Intelligence (AI) and data mining not only provide a better understanding of how real-world systems function, but they also enable us to predict system behavior before a system is actually built. They can also accurately analyze systems under varying operating conditions. This book provides comprehensive, state-of-the-art coverage of all the important aspects of modeling and simulating both physical and conceptual systems. Various real-life examples show how simulation plays a key role in understanding real-world systems. We also explained how to effectively use AI and Data Mining techniques to successfully apply the modeling and simulation techniques presented.
After introducing the underlying philosophy of systems, the book offers step-by-step procedures for modeling with practical examples and coding different types of systems using modeling techniques, such as the Pattern Recognition, Automatic Threat detection, Automatic problem solving, etc.
Preparing both undergraduate and graduate students for advanced modeling and simulation courses, this text helps them carry out effective simulation studies. In addition, graduate students should be able to comprehend and conduct AI and Data Mining research after completing this book.
This book is organized into fifteen chapters. In Chapter 1, this Chapter discusses about the cyber security needs that can be addressed by AI techniques. It talks about the traditional approach and how AI can be used to modify the multilayered security mechanism used in companies today. Here we propose a system that adds additional layer of security in order to detect any unwanted intrusion. The ever-expanding danger of digital assaults, cybercrimes, and malware attacks has grown exponentially with evolution of artificial intelligence. Conventional ways of cyber-attacks have now taken a turning point, consequently, the attackers resort to more intelligent ways.
In Chapter 2, we have tried to show the power of intrusion detection is the most important application of data mining by applying different data mining techniques to detect it effectively and report the same in actual time so that essential and required arrangements can be made to stop the efforts made by the trespassery.
In Chapter 3, we have explained about how Artificial Intelligence (AI) is a popular expression in the digital world. It is as yet a creating science in various features as indicated by the difficulties tossed by 21st century. Usage of artificial intelligence has gotten undefined from human life. Nowadays one can't imagine a world without AI as it has a ton of gigantic impact on human life. The essential objective of artificial intelligence is to develop the advancement based activities which addresses the human data in order to handle issues. Basically artificial intelligence is examination of how an individual think, work, learn and pick in any circumstance of life, whether or not it may be related to basic reasoning or learning new things or thinking equitably or to appear at an answer, etc.
In Chapter 4, we have explained further proposed a botnet identification version using optics algorithm that hopes to effectively discover botnets and perceive the type botnet detected by way of addition of latest feature; incorporation of changed traces to pinpoint supply IP of bot master, identification of existence of the kind of services the botnets have get right of entry to to are areas the proposed solution will cater for.
In Chapter 5, we have explained about models basically 'learns' from experience with respect to some task and are capable of finding 'commonality' in many different observations. This study discusses various methods of spam filtering using existing Artificial Intelligence techniques and compares their strengths and limitations.
In Chapter 6, we have explained about how as artificial intelligence people in general to improve, there are risks associated with their utilization, set up in functioning frameworks, tools, calculations, framework the executives, morals and duty, and privacy. The study focuses around the risks and threats of computerized reasoning and how AI can help comprehend network safety or areas of cyber security issues.
In Chapter 7, we have explained about problem to make privacy in multi-tenant in the single framework. For that problem we use the artificial intelligence concept to improve the security and privacy concept in multitenant based system. Using Artificial intelligence the privacy and security concept make strong because in artificial intelligence work as intelligent human or animal mind it make maximum changes to fulfill the requirement of the concept to achieve the goal. In this chapter describes the issues of privacy and security problems in multi tenancy.
In Chapter 8, we have provided detailed explanations of a novel approach for biometric recognition has been introduced in which the application of ILBP (Improved Local Binary Pattern) for facial feature detection is discussed which generates the improved features for the facial pattern. It allows only authenticated user to access a system which is better than previous algorithms. Previous research for face detection shows many demerits in terms of false acceptance and rejection rates. In this paper, the extraction of facial features is done from static and dynamic frames using the Haar cascade algorithm.
In Chapter 9, we have explained about a the developed system consists of a climbing robot, camera for image capturing, IoT modules for transmitting images to cloud, image processing platform, and artificial neural network module intended for decision making. Climbing robot holds the cable with the grooved wheels along with the auto trigger camera and the IoT module. For inspection, the robot ascends along the cables continuously and acquires images of various segments of the cable. Then the captured images have been send to the cloud storage through IoT system. The stored images have been retrieved and their sizes have been reduced through the image processing techniques.
In Chapter 10, we have a digital security threats results from the character of those omnipresent and at times over the top interchanges interconnections. Digital security isn't one aspect, yet rather it's a gaggle of profoundly various issues mentions various arrangements of threats. An Advance Cyber Security System utilizing emblematic rationale might be a framework that comprises of a standard safe and an instrument for getting to and running the standards. The vault is ordinarily built with a lot of related standard sets. Fuzzy improvement manages finding the estimations of information boundaries of a luxurious recreated framework which winds up in wanted yield.
In Chapter 11, the goal of current chapter is to analyze cyber threats and to demonstrate how artificial intelligence and data mining approaches can be effective to fix cyber-attack issues. The field of artificial intelligence has been increasingly playing a vital role in analyzing cyber threat and improving cyber security as well as safety. Mainly three aspects are discussed in this chapter. First the process of cyber-attack detection which will help to analyses and classify cyber incident, Second task is forecasting upcoming cyber-attack and to control the cyber terrorism. Finally the chapter focus on theoretical background and practical usability of artificial intelligence with data mining approaches for addressing above detection and prediction.
In Chapter 12, this chapter explores the modern intrusion detection with a distinctive determination perspective of data mining. This discussion focuses on major facets of intrusion detection strategy that is misuse detection. Below content focuses on, to identify attacks, information or data which is present on the network using C4.5 algorithm, which is type of decision tree technique and also it helps to enhance the IDS system to recognize types of attacks in network. For this attack detection, KDD-99 dataset is used, contains several features and different class of general and attack type data.
In Chapter 13, in this current research, firefly algorithm has been used for optimizing maize crop yield by considering the various constraints and risks. This research investigates the development of new firefly algorithm module for predicting the optimal climatic conditions and predicts the crop cultivation output. As the pre-processing, the maize crop cultivation data for 96 months have been collected and provided as response to Minitab software to formulate the relational equation. The collected data have been stored in the cloud using IoT and the cloud has to be updated periodically for obtaining the accurate results from the algorithm.
In Chapter 14, gestures are of two types as: static and dynamic sequences, this is where vision based techniques plays a vital role. The survey on the research study on the vision-based gesture recognition approaches have been briefed in this paper. Challenges in all perspective of recognition of gestures using images are detailed. A systematic review has been conducted over 100 papers and narrowed down into 60 papers on summarized. The foremost motive of this paper is to provide a strong foundation on vision based recognition and apply this for solutions in medical and engineering fields. Outlines gaps & current trends to motivate researchers to improve their contribution.
In Chapter 15, we will cover a examine of diverse thoughts, attempts, efficiency and different studies trends in junk mail filtering. The history observe explains the packages of...
