Machine learning techniques and analytics for cloud security / edited by Rajdeep Chakraborty, Anupam Ghosh and Jyotsna Kumar Mandal.

Includes bibliographical references and index.

Table of Contents

Contents

Preface

Part I: Conceptual Aspects on Cloud and Applications of Machine Learning 1

1 Hybrid Cloud: A New Paradigm in Cloud Computing 3
Moumita Deb and Abantika Choudhury

1.1 Introduction 3

1.2 Hybrid Cloud 5

1.2.1 Architecture 6

1.2.2 Why Hybrid Cloud is Required? 6

1.2.3 Business and Hybrid Cloud 7

1.2.4 Things to Remember When Deploying Hybrid Cloud 8

1.3 Comparison Among Different Hybrid Cloud Providers 9

1.3.1 Cloud Storage and Backup Benefits 11

1.3.2 Pros and Cons of Different Service Providers 11

1.3.2.1 AWS Outpost 12

1.3.2.2 Microsoft Azure Stack 12

1.3.2.3 Google Cloud Anthos 12

1.3.3 Review on Storage of the Providers 13

1.3.3.1 AWS Outpost Storage 13

1.3.3.2 Google Cloud Anthos Storage 13

1.3.4 Pricing 15

1.4 Hybrid Cloud in Education 15

1.5 Significance of Hybrid Cloud Post-Pandemic 15

1.6 Security in Hybrid Cloud 16

1.6.1 Role of Human Error in Cloud Security 18

1.6.2 Handling Security Challenges 18

1.7 Use of AI in Hybrid Cloud 19

1.8 Future Research Direction 21

1.9 Conclusion 22

References 22

xix

v

2 Recognition of Differentially Expressed Glycan Structure of H1N1 Virus Using Unsupervised Learning Framework 25
Shillpi Mishrra

2.1 Introduction 25

2.2 Proposed Methodology 27

2.3 Result 28

2.3.1 Description of Datasets 29

2.3.2 Analysis of Result 29

2.3.3 Validation of Results 31

2.3.3.1 T-Test (Statistical Validation) 31

2.3.3.2 Statistical Validation 33

2.3.4 Glycan Cloud 37

2.4 Conclusions and Future Work 38

References 39

3 Selection of Certain Cancer Mediating Genes Using a Hybrid Model Logistic Regression Supported by Principal Component Analysis (PC-LR) 41
Subir Hazra, Alia Nikhat Khurshid and Akriti

3.1 Introduction 41

3.2 Related Methods 44

3.3 Methodology 46

3.3.1 Description 47

3.3.2 Flowchart 49

3.3.3 Algorithm 49

3.3.4 Interpretation of the Algorithm 50

3.3.5 Illustration 50

3.4 Result 51

3.4.1 Description of the Dataset 51

3.4.2 Result Analysis 51

3.4.3 Result Set Validation 52

3.5 Application in Cloud Domain 56

3.6 Conclusion 58

References 59

Part II: Cloud Security Systems Using Machine Learning Techniques 61

4 Cost-Effective Voice-Controlled Real-Time Smart Informative Interface Design With Google Assistance Technology 63
Soumen Santra, Partha Mukherjee and Arpan Deyasi

4.1 Introduction 64

4.2 Home Automation System 65

4.2.1 Sensors 65

4.2.2 Protocols 66

4.2.3 Technologies 66

4.2.4 Advantages 67

4.2.5 Disadvantages 67

4.3 Literature Review 67

4.4 Role of Sensors and Microcontrollers in Smart Home Design 68

4.5 Motivation of the Project 70

4.6 Smart Informative and Command Accepting Interface 70

4.7 Data Flow Diagram 71

4.8 Components of Informative Interface 72

4.9 Results 73

4.9.1 Circuit Design 73

4.9.2 LDR Data 76

4.9.3 API Data 76

4.10 Conclusion 78

4.11 Future Scope 78

References 78

5 Symmetric Key and Artificial Neural Network With Mealy Machine: A Neoteric Model of Cryptosystem for Cloud Security 81
Anirban Bhowmik, Sunil Karforma and Joydeep Dey

5.1 Introduction 81

5.2 Literature Review 85

5.3 The Problem 86

5.4 Objectives and Contributions 86

5.5 Methodology 87

5.6 Results and Discussions 91

5.6.1 Statistical Analysis 93

5.6.2 Randomness Test of Key 94

5.6.3 Key Sensitivity Analysis 95

5.6.4 Security Analysis 96

5.6.5 Dataset Used on ANN 96

5.6.6 Comparisons 98

5.7 Conclusions 99

References 99

6 An Efficient Intrusion Detection System on Various Datasets Using Machine Learning Techniques 103
Debraj Chatterjee

6.1 Introduction 103

6.2 Motivation and Justification of the Proposed Work 104

6.3 Terminology Related to IDS 105

6.3.1 Network 105

6.3.2 Network Traffic 105

6.3.3 Intrusion 106

6.3.4 Intrusion Detection System 106

6.3.4.1 Various Types of IDS 108

6.3.4.2 Working Methodology of IDS 108

6.3.4.3 Characteristics of IDS 109

6.3.4.4 Advantages of IDS 110

6.3.4.5 Disadvantages of IDS 111

6.3.5 Intrusion Prevention System (IPS) 111

6.3.5.1 Network-Based Intrusion Prevention System (NIPS) 111

6.3.5.2 Wireless Intrusion Prevention System (WIPS) 112

6.3.5.3 Network Behavior Analysis (NBA) 112

6.3.5.4 Host-Based Intrusion Prevention System (HIPS) 112

6.3.6 Comparison of IPS With IDS/Relation Between IDS and IPS 112

6.3.7 Different Methods of Evasion in Networks 113

6.4 Intrusion Attacks on Cloud Environment 114

6.5 Comparative Studies 116

6.6 Proposed Methodology 121

6.7 Result 122

6.8 Conclusion and Future Scope 125

References 126

7 You Are Known by Your Mood: A Text-Based Sentiment Analysis for Cloud Security 129
Abhijit Roy and Parthajit Roy

7.1 Introduction 129

7.2 Literature Review 131

7.3 Essential Prerequisites 133

7.3.1 Security Aspects 133

7.3.2 Machine Learning Tools 135

7.3.2.1 Naïve Bayes Classifier 135

7.3.2.2 Artificial Neural Network 136

7.4 Proposed Model 136

7.5 Experimental Setup 138

7.6 Results and Discussions 139

7.7 Application in Cloud Security 142

7.7.1 Ask an Intelligent Security Question 142

7.7.2 Homomorphic Data Storage 142

7.7.3 Information Diffusion 144

7.8 Conclusion and Future Scope 144

References 145

8 The State-of-the-Art in Zero-Knowledge Authentication Proof for Cloud 149
Priyanka Ghosh

8.1 Introduction 149

8.2 Attacks and Countermeasures 153

8.2.1 Malware and Ransomware Breaches 154

8.2.2 Prevention of Distributing Denial of Service 154

8.2.3 Threat Detection 154

8.3 Zero-Knowledge Proof 154

8.4 Machine Learning for Cloud Computing 156

8.4.1 Types of Learning Algorithms 156

8.4.1.1 Supervised Learning 156

8.4.1.2 Supervised Learning Approach 156

8.4.1.3 Unsupervised Learning 157

8.4.2 Application on Machine Learning for Cloud Computing 157

8.4.2.1 Image Recognition 157

8.4.2.2 Speech Recognition 157

8.4.2.3 Medical Diagnosis 158

8.4.2.4 Learning Associations 158

8.4.2.5 Classification 158

8.4.2.6 Prediction 158

8.4.2.7 Extraction 158

8.4.2.8 Regression 158

8.4.2.9 Financial Services 159

8.5 Zero-Knowledge Proof: Details 159

8.5.1 Comparative Study 159

8.5.1.1 Fiat-Shamir ZKP Protocol 159

8.5.2 Diffie-Hellman Key Exchange Algorithm 161

8.5.2.1 Discrete Logarithm Attack 161

8.5.2.2 Man-in-the-Middle Attack 162

8.5.3 ZKP Version 1 162

8.5.4 ZKP Version 2 162

8.5.5 Analysis 164

8.5.6 Cloud Security Architecture 166

8.5.7 Existing Cloud Computing Architectures 167

8.5.8 Issues With Current Clouds 167

8.6 Conclusion 168

References 169

9 A Robust Approach for Effective Spam Detection Using Supervised Learning Techniques 171
Amartya Chakraborty, Suvendu Chattaraj, Sangita Karmakar and Shillpi Mishrra

9.1 Introduction 171

9.2 Literature Review 173

9.3 Motivation 174

9.4 System Overview 175

9.5 Data Description 176

9.6 Data Processing 176

9.7 Feature Extraction 178

9.8 Learning Techniques Used 179

9.8.1 Support Vector Machine 179

9.8.2 k-Nearest Neighbors 180

9.8.3 Decision Tree 180

9.8.4 Convolutional Neural Network 180

9.9 Experimental Setup 182

9.10 Evaluation Metrics 183

9.11 Experimental Results 185

9.11.1 Observations in Comparison With State-of-the-Art 187

9.12 Application in Cloud Architecture 188

9.13 Conclusion 189

References 190

10 An Intelligent System for Securing Network From Intrusion Detection and Prevention of Phishing Attack Using Machine Learning Approaches 193
Sumit Banik, Sagar Banik and Anupam Mukherjee

10.1 Introduction 193

10.1.1 Types of Phishing 195

10.1.1.1 Spear Phishing 195

10.1.1.2 Whaling 195

10.1.1.3 Catphishing and Catfishing 195

10.1.1.4 Clone Phishing 196

10.1.1.5 Voice Phishing 196

10.1.2 Techniques of Phishing 196

10.1.2.1 Link Manipulation 196

10.1.2.2 Filter Evasion 196

10.1.2.3 Website Forgery 196

10.1.2.4 Covert Redirect 197

10.2 Literature Review 197

10.3 Materials and Methods 199

10.3.1 Dataset and Attributes 199

10.3.2 Proposed Methodology 199

10.3.2.1 Logistic Regression 202

10.3.2.2 Naïve Bayes 202

10.3.2.3 Support Vector Machine 203

10.3.2.4 Voting Classification 203

10.4 Result Analysis 204

10.4.1 Analysis of Different Parameters for ML Models 204

10.4.2 Predictive Outcome Analysis in Phishing URLs Dataset 205

10.4.3 Analysis of Performance Metrics 206

10.4.4 Statistical Analysis of Results 210

‌0.4.4. 1 ANOVA: Two-Factor Without Replication 210

10.4.4.2 ANOVA: Single Factor 210

10.5 Conclusion 210

References 211

Part III: Cloud Security Analysis Using Machine Learning Techniques 213

11 Cloud Security Using Honeypot Network and Blockchain: A Review 215
Smarta Sangui * and Swarup Kr Ghosh

11.1 Introduction 215

11.2 Cloud Computing Overview 216

11.2.1 Types of Cloud Computing Services 216

11.2.1.1 Software as a Service 216

11.2.1.2 Infrastructure as a Service 218

11.2.1.3 Platform as a Service 218

11.2.2 Deployment Models of Cloud Computing 218

11.2.2.1 Public Cloud 218

11.2.2.2 Private Cloud 218

11.2.2.3 Community Cloud 219

11.2.2.4 Hybrid Cloud 219

11.2.3 Security Concerns in Cloud Computing 219

11.2.3.1 Data Breaches 219

11.2.3.2 Insufficient Change Control and Misconfiguration 219

11.2.3.3 Lack of Strategy and Security Architecture 220

11.2.3.4 Insufficient Identity, Credential, Access, and Key Management 220

11.2.3.5 Account Hijacking 220

11.2.3.6 Insider Threat 220

11.2.3.7 Insecure Interfaces and APIs 220

11.2.3.8 Weak Control Plane 221

11.3 Honeypot System 221

11.3.1 VM (Virtual Machine) as Honeypot in the Cloud 221

11.3.2 Attack Sensing and Analyzing Framework 222

11.3.3 A Fuzzy Technique Against Fingerprinting Attacks 223

11.3.4 Detecting and Classifying Malicious Access 224

11.3.5 A Bayesian Defense Model for Deceptive Attack 224

11.3.6 Strategic Game Model for DDoS Attacks in Smart Grid 226

11.4 Blockchain 227

11.4.1 Blockchain-Based Encrypted Cloud Storage 228

11.4.2 Cloud-Assisted EHR Sharing via Consortium Blockchain 229

11.4.3 Blockchain-Secured Cloud Storage 230

11.4.4 Blockchain and Edge Computing–Based Security Architecture 230

11.4.5 Data Provenance Architecture in Cloud Ecosystem Using Blockchain 231

11.6 Comparative Analysis 233

11.7 Conclusion 233

References 234

12 Machine Learning–Based Security in Cloud Database—A Survey 239
Utsav Vora, Jayleena Mahato, Hrishav Dasgupta, Anand Kumar and Swarup Kr Ghosh

12.1 Introduction 239

12.2 Security Threats and Attacks 241

12.3 Dataset Description 244

12.3.1 NSL-KDD Dataset 244

12.3.2 UNSW-NB15 Dataset 244

12.4 Machine Learning for Cloud Security 245

12.4.1 Supervised Learning Techniques 245

12.4.1.1 Support Vector Machine 245

12.4.1.2 Artificial Neural Network 247

12.4.1.3 Deep Learning 249

12.4.1.4 Random Forest 250

12.4.2 Unsupervised Learning Techniques 251

12.4.2.1 K-Means Clustering 252

12.4.2.2 Fuzzy C-Means Clustering 253

12.4.2.3 Expectation-Maximization Clustering 253

12.4.2.4 Cuckoo Search With Particle Swarm Optimization (PSO) 254

12.4.3 Hybrid Learning Techniques 256

12.4.3.1 HIDCC: Hybrid Intrusion Detection Approach in Cloud Computing 256

12.4.3.2 Clustering-Based Hybrid Model in Deep Learning Framework 257

12.4.3.3 K-Nearest Neighbor–Based Fuzzy C-Means Mechanism 258

12.4.3.4 K-Means Clustering Using Support Vector Machine 260

12.4.3.5 K-Nearest Neighbor–Based Artificial Neural Network Mechanism 260

12.4.3.6 Artificial Neural Network Fused With Support Vector Machine 261

12.4.3.7 Particle Swarm Optimization–Based Probabilistic Neural Network 261

12.5 Comparative Analysis 262

12.6 Conclusion 264

References 267

13 Machine Learning Adversarial Attacks: A Survey Beyond 271
Chandni Magoo and Puneet Garg

13.1 Introduction 271

13.2 Adversarial Learning 272

13.2.1 Concept 272

13.3 Taxonomy of Adversarial Attacks 273

13.3.1 Attacks Based on Knowledge 273

13.3.1.1 Black Box Attack (Transferable Attack) 273

13.3.1.2 White Box Attack 274

13.3.2 Attacks Based on Goals 275

13.3.2.1 Target Attacks 275

13.3.2.2 Non-Target Attacks 275

13.3.3 Attacks Based on Strategies 275

13.3.3.1 Poisoning Attacks 275

13.3.3.2 Evasion Attacks 276

13.3.4 Textual-Based Attacks (NLP) 276

13.3.4.1 Character Level Attacks 276

13.3.4.2 Word-Level Attacks 276

13.3.4.3 Sentence-Level Attacks 276

13.4 Review of Adversarial Attack Methods 276

13.4.1 L-bfgs 277

13.4.2 Feedforward Derivation Attack (Jacobian Attack) 277

13.4.3 Fast Gradient Sign Method 278

13.4.4 Methods of Different Text-Based Adversarial Attacks 278

13.4.5 Adversarial Attacks Methods Based on Language Models 284

13.4.6 Adversarial Attacks on Recommender Systems 284

13.4.6.1 Random Attack 284

13.4.6.2 Average Attack 286

13.4.6.3 Bandwagon Attack 286

13.4.6.4 Reverse Bandwagon Attack 286

13.5 Adversarial Attacks on Cloud-Based Platforms 287

13.6 Conclusion 288

References 288

14 Protocols for Cloud Security 293
Weijing You and Bo Chen

14.1 Introduction 293

14.2 System and Adversarial Model 295

14.2.1 System Model 295

14.2.2 Adversarial Model 295

14.3 Protocols for Data Protection in Secure Cloud Computing 296

14.3.1 Homomorphic Encryption 297

14.3.2 Searchable Encryption 298

14.3.3 Attribute-Based Encryption 299

14.3.4 Secure Multi-Party Computation 300

14.4 Protocols for Data Protection in Secure Cloud Storage 301

14.4.1 Proofs of Encryption 301

14.4.2 Secure Message-Locked Encryption 303

14.4.3 Proofs of Storage 303

14.4.4 Proofs of Ownership 305

14.4.5 Proofs of Reliability 306

14.5 Protocols for Secure Cloud Systems 309

14.6 Protocols for Cloud Security in the Future 309

14.7 Conclusion 310

References 311

Part IV: Case Studies Focused on Cloud Security 313

15 A Study on Google Cloud Platform (GCP) and Its Security 315
Agniswar Roy, Abhik Banerjee and Navneet Bhardwaj

15.1 Introduction 315

15.1.1 Google Cloud Platform Current Market Holding 316

15.1.1.1 The Forrester Wave 317

15.1.1.2 Gartner Magic Quadrant 317

15.1.2 Google Cloud Platform Work Distribution 317

15.1.2.1 SaaS 318

15.1.2.2 PaaS 318

15.1.2.3 IaaS 318

15.1.2.4 On-Premise 318

15.2 Google Cloud Platform’s Security Features Basic Overview 318

15.2.1 Physical Premises Security 319

15.2.2 Hardware Security 319

15.2.3 Inter-Service Security 319

15.2.4 Data Security 320

15.2.5 Internet Security 320

15.2.6 In-Software Security 320

15.2.7 End User Access Security 321

15.3 Google Cloud Platform’s Architecture 321

15.3.1 Geographic Zone 321

15.3.2 Resource Management 322

15.3.2.1 Iam 322

15.3.2.2 Roles 323

15.3.2.3 Billing 323

15.4 Key Security Features 324

15.4.1 Iap 324

15.4.2 Compliance 325

15.4.3 Policy Analyzer 326

15.4.4 Security Command Center 326

15.4.4.1 Standard Tier 326

15.4.4.2 Premium Tier 326

15.4.5 Data Loss Protection 329

15.4.6 Key Management 329

15.4.7 Secret Manager 330

15.4.8 Monitoring 330

15.5 Key Application Features 330

15.5.1 Stackdriver (Currently Operations) 330

15.5.1.1 Profiler 330

15.5.1.2 Cloud Debugger 330

15.5.1.3 Trace 331

15.5.2 Network 331

15.5.3 Virtual Machine Specifications 332

15.5.4 Preemptible VMs 332

15.6 Computation in Google Cloud Platform 332

15.6.1 Compute Engine 332

15.6.2 App Engine 333

15.6.3 Container Engine 333

15.6.4 Cloud Functions 333

15.7 Storage in Google Cloud Platform 333

15.8 Network in Google Cloud Platform 334

15.9 Data in Google Cloud Platform 334

15.10 Machine Learning in Google Cloud Platform 335

15.11 Conclusion 335

References 337

16 Case Study of Azure and Azure Security Practices 339
Navneet Bhardwaj, Abhik Banerjee and Agniswar Roy

16.1 Introduction 339

16.1.1 Azure Current Market Holding 340

16.1.2 The Forrester Wave 340

16.1.3 Gartner Magic Quadrant 340

16.2 Microsoft Azure—The Security Infrastructure 341

16.2.1 Azure Security Features and Tools 341

16.2.2 Network Security 342

16.3 Data Encryption 342

16.3.1 Data Encryption at Rest 342

16.3.2 Data Encryption at Transit 342

16.3.3 Asset and Inventory Management 343

16.3.4 Azure Marketplace 343

16.4 Azure Cloud Security Architecture 344

16.4.1 Working 344

16.4.2 Design Principles 344

16.4.2.1 Alignment of Security Policies 344

16.4.2.2 Building a Comprehensive Strategy 345

16.4.2.3 Simplicity Driven 345

16.4.2.4 Leveraging Native Controls 345

16.4.2.5 Identification-Based Authentication 345

16.4.2.6 Accountability 345

16.4.2.7 Embracing Automation 345

16.4.2.8 Stress on Information Protection 345

16.4.2.9 Continuous Evaluation 346

16.4.2.10 Skilled Workforce 346

16.5 Azure Architecture 346

16.5.1 Components 346

16.5.1.1 Azure Api Gateway 346

16.5.1.2 Azure Functions 346

16.5.2 Services 347

16.5.2.1 Azure Virtual Machine 347

16.5.2.2 Blob Storage 347

16.5.2.3 Azure Virtual Network 348

16.5.2.4 Content Delivery Network 348

16.5.2.5 Azure SQL Database 349

16.6 Features of Azure 350

16.6.1 Key Features 350

16.6.1.1 Data Resiliency 350

16.6.1.2 Data Security 350

16.6.1.3 BCDR Integration 350

16.6.1.4 Storage Management 351

16.6.1.5 Single Pane View 351

16.7 Common Azure Security Features 351

16.7.1 Security Center 351

16.7.2 Key Vault 351

16.7.3 Azure Active Directory 352

16.7.3.1 Application Management 352

16.7.3.2 Conditional Access 352

16.7.3.3 Device Identity Management 352

​16.7.3. 4 Identity Protection 353

16.7.3.5 Azure Sentinel 353

16.7.3.6 Privileged Identity Management 354

16.7.3.7 Multifactor Authentication 354

16.7.3.8 Single Sign On 354

16.8 Conclusion 355

References 355

17 Nutanix Hybrid Cloud From Security Perspective 357
Abhik Banerjee, Agniswar Roy, Amar Kalvikatte and Navneet Bhardwaj

17.1 Introduction 357

17.2 Growth of Nutanix 358

17.2.1 Gartner Magic Quadrant 358

17.2.2 The Forrester Wave 358

17.2.3 Consumer Acquisition 359

17.2.4 Revenue 359

17.3 Introductory Concepts 361

17.3.1 Plane Concepts 361

17.3.1.1 Control Plane 361

17.3.1.2 Data Plane 361

17.3.2 Security Technical Implementation Guides 362

17.3.3 SaltStack and SCMA 362

17.4 Nutanix Hybrid Cloud 362

17.4.1 Prism 362

17.4.1.1 Prism Element 363

17.4.1.2 Prism Central 364

17.4.2 Acropolis 365

17.4.2.1 Distributed Storage Fabric 365

17.4.2.2 Ahv 367

17.5 Reinforcing AHV and Controller VM 367

17.6 Disaster Management and Recovery 368

17.6.1 Protection Domains and Consistent Groups 368

17.6.2 Nutanix DSF Replication of OpLog 369

17.6.3 DSF Snapshots and VmQueisced Snapshot Service 370

17.6.4 Nutanix Cerebro 370

17.7 Security and Policy Management on Nutanix Hybrid Cloud 371

17.7.1 Authentication on Nutanix 372

17.7.2 Nutanix Data Encryption 372

17.7.3 Security Policy Management 373

17.7.3.1 Enforcing a Policy 374

17.7.3.2 Priority of a Policy 374

17.7.3.3 Automated Enforcement 374

17.8 Network Security and Log Management 374

17.8.1 Segmented and Unsegmented Network 375

17.9 Conclusion 376

References 376

Part V: Policy Aspects 379

18 A Data Science Approach Based on User Interactions to Generate Access Control Policies for Large Collections of Documents 381
Jedidiah Yanez-Sierra, Arturo Diaz-Perez and Victor Sosa-Sosa

18.1 Introduction 381

18.2 Related Work 383

18.3 Network Science Theory 384

18.4 Approach to Spread Policies Using Networks Science 387

18.4.1 Finding the Most Relevant Spreaders 388

18.4.1.1 Weighting Users 389

18.4.1.2 Selecting the Top � Spreaders 390

18.4.2 Assign and Spread the Access Control Policies 390

18.4.2.1 Access Control Policies 391

18.4.2.2 Horizontal Spreading 391

18.4.2.3 Vertical Spreading (Bottom-Up) 392

18.4.2.4 Policies Refinement 395

18.4.3 Structural Complexity Analysis of CP-ABE Policies 395

18.4.3.1 Assessing the WSC for ABE Policies 396

18.4.3.2 Assessing the Policies Generated in the Spreading Process 397

18.4.4 Effectiveness Analysis 398

18.4.4.1 Evaluation Metrics 399

18.4.4.2 Adjusting the Interaction Graph to Assess Policy Effectiveness 400

18.4.4.3 Method to Complement the User Interactions (Synthetic Edges Generation) 400

18.4.5 Measuring Policy Effectiveness in the User Interaction Graph 403

18.4.5.1 Simple Node-Based Strategy 403

18.4.5.2 Weighted Node-Based Strategy 404

18.5 Evaluation 405

18.5.1 Dataset Description 405

18.5.2 Results of the Complexity Evaluation 406

18.5.3 Effectiveness Results From the Real Edges 407

18.5.4 Effectiveness Results Using Real and Synthetic Edges 408

18.5.4.1 Results of the Effectiveness Metrics for the Enhanced G + Graph 410

18.6 Conclusions 413

References 414

19 AI, ML, & Robotics in iSchools: An Academic Analysis for an Intelligent Societal Systems 417
P. K. Paul

19.1 Introduction 417

19.2 Objective 419

19.3 Methodology 420

19.3.1 iSchools, Technologies, and Artificial Intelligence, ML, and Robotics 420

19.4 Artificial Intelligence, ML, and Robotics: An Overview 427

19.5 Artificial Intelligence, ML, and Robotics as an Academic Program: A Case on iSchools—North American Region 428

19.6 Suggestions 431

19.7 Motivation and Future Works 435

19.8 Conclusion 435

References 436

Index 439

This book covers new methods, surveys, case studies, and policy with almost all machine learning techniques and analytics for cloud security solutions. The aim of Machine Learning Techniques and Analytics for Cloud Security is to integrate machine learning approaches to meet various analytical issues in cloud security. Cloud security with ML has long-standing challenges that require methodological and theoretical handling. The conventional cryptography approach is less applied in resource-constrained devices. To solve these issues, the machine learning approach may be effectively used in providing security to the vast growing cloud environment. Machine learning algorithms can also be used to meet various cloud security issues, such as effective intrusion detection systems, zero-knowledge authentication systems, measures for passive attacks, protocols design, privacy system designs, applications, and many more. The book also contains case studies/projects outlining how to implement various security features using machine learning algorithms and analytics on existing cloud-based products in public, private and hybrid cloud respectively.

About the Author

Rajdeep Chakraborty obtained his PhD in CSE from the University of Kalyani. He is currently an assistant professor in the Department of Computer Science and Engineering, Netaji Subhash Engineering College, Garia, Kolkata, India. He has several publications in reputed international journals and conferences and has authored a book on hardware cryptography. His field of interest is mainly in cryptography and computer security.

Anupam Ghosh obtained his PhD in Engineering from Jadavpur University. He is currently a professor in the Department of Computer Science and Engineering, Netaji Subhash Engineering College, Kolkata. He has published more than 80 papers in reputed international journals and conferences. His field of interest is mainly in AI, machine learning, deep learning, image processing, soft computing, bioinformatics, IoT, data mining.

Jyotsna Kumar Mandal obtained his PhD in CSE from Jadavpur University He has more than 450 publications in reputed international journals and conferences. His field of interest is mainly in coding theory, data and network security, remote sensing & GIS-based applications, data compression error corrections, information security, watermarking, steganography and document authentication, image processing, visual cryptography, MANET, wireless and mobile computing/security, unify computing, chaos theory, and applications.