Intelligent Decision Support Systems For Sustainable Computing: Paradigms And Applications (studies In Computational Intelligence)

This unique book dicusses the latest research, innovative ideas, challenges and computational intelligence (CI) solutions in sustainable computing. It presents novel, in-depth fundamental research on achieving a sustainable lifestyle for society, either from a methodological or from an application perspective. Sustainable computing has expanded to become a significant research area covering the fields of computer science and engineering, electrical engineering and other engineering disciplines, and there has been an increase in the amount of literature on aspects sustainable computing such as energy efficiency and natural resources conservation that emphasizes the role of ICT (information and communications technology) in achieving system design and operation objectives. The energy impact/design of more efficient IT infrastructures is a key challenge in realizing new computing paradigms. The book explores the uses of computational intelligence (CI) techniques for intelligent decision support that can be exploited to create effectual computing systems, and addresses sustainability problems in computing and information processing environments and technologies at the different levels of CI paradigms. An excellent guide to surveying the state of the art in computational intelligence applied to challenging real-world problems in sustainable computing, it is intended for scientists, practitioners, researchers and academicians dealing with the new challenges and advances in area. Foreword 6 Preface 8 Need for a Book on the Proposed Topics 9 Organization of the Book 9 Audience 11 Acknowledgements 13 Contents 14 1 Intelligent Decision Support Systems for Sustainable Computing 16 Abstract 16 1 Introduction 17 2 Intelligent Decision Support Computational Intelligence Paradigms 17 3 CI Paradigms for Sustainable Applications 19 4 Importance of CI in Sustainable Computing Research 19 5 Conclusion 20 References 21 A Genetic Algorithm Based Efficient Static Load Distribution Strategy for Handling Large-Scale Workloads on Sustainable Computing Systems 22 1 Introduction 22 2 Availability-Aware Scheduling Model 25 2.1 Problem Description 25 2.2 A Novel Scheduling Model for Sustainable Computing 28 3 Algorithm PAA-GA Based Global Optimization Strategy 29 3.1 Encoding Scheme 30 3.2 Crossover Operators 30 3.3 Mutation Operator 33 3.4 Repair Operator 34 3.5 Local Search 35 3.6 Framework of Algorithm PAA-GA 38 4 Experimental Results and Analysis 38 4.1 Evaluating the Correctness of PAA-GA 38 4.2 Evaluating the Performance of PAA-GA 42 5 Conclusions 44 References 44 3 Efficiency in Energy Decision Support Systems Using Soft Computing Techniques 47 Abstract 47 1 Introduction 48 2 Literature Review 49 3 Materials and Methods 52 3.1 DSS Logical Architecture 52 3.2 Adaptive Neuro Fuzzy Inference Systems 53 3.3 Artificial Neural Networks (ANN) 56 3.4 Evolutionary Fuzzy Cognitive Maps 56 3.4.1 Econometric Model for Natural Gas Forecasting 58 4 Results and Discussion 59 5 Conclusions 63 References 64 4 Computational Intelligence Based Heuristic Approach for Maximizing Energy Efficiency in Internet of Things 67 Abstract 67 1 Introduction 68 2 Related Works 70 3 System Model and Assumptions 71 3.1 Application Model and Mobile Cloud Model 71 4 Task Scheduling Based on Multi-objective Particle Swarm Optimization 73 4.1 Multi-objective Optimization 74 4.2 Particle Swarm Optimization (PSO) 74 5 Proposed Work 76 5.1 First Phase: Initial Schedule 76 5.1.1 An Example 77 5.2 Second Phase: MMOPSO Algorithm 77 6 Performance Evaluations 80 6.1 Experimental Setup 81 6.2 Performance Metrics 82 6.3 Simulation Results 83 7 Conclusion and Future Work 87 References 87 5 Distributed Algorithm with Inherent Intelligence for Multi-cloud Resource Provisioning 91 Abstract 91 1 Introduction 92 2 Background in Resource Management and Upcoming Challenge 94 3 Ranked System in Multi-cloud 94 3.1 Utility Functions in Ranked System 98 4 Mathematical Formulation of the Algorithm 100 4.1 Dynamic Distributed Resource Provisioning Approach 103 5 Proposed Algorithm for Resource Scheduling 105 5.1 Simulation Results 106 5.2 Implementation of Spot Instances with the Logarithmic Function 109 6 Conclusion 111 References 111 6 Parameter Optimization Methods Based on Computational Intelligence Techniques in Context of Sustainable Computing 114 Abstract 114 1 Introduction 115 2 Parameter Optimization Taxonomy 116 3 Parameter Optimization Technique 117 3.1 Meta-optimization Methods 118 3.2 Self Tuning 120 3.3 Bayesian Case Based Method 121 3.4 Bi-level Optimization Approach 121 3.5 Additive Procedures 122 4 Algorithm Quality: Performance and Robustness 123 4.1 Performance Measures 123 4.2 Robustness 123 5 Conclusion and Discussion 124 References 124 7 The Maximum Power Point Tracking Using Fuzzy Logic Algorithm for DC Motor Based Conveyor System 127 Abstract 127 1 Introduction 128 2 Maximum Power Point Tracking System 130 2.1 Perturb and Observe (P&O) Algorithm 131 2.2 Incremental Conductance (IC) Algorithm 132 3 Fuzzy Logic Based MPPT Algorithm 135 3.1 Design of Fuzzy Logic Controller 135 4 Design of DC Motor Drive Circuits and Conveyor System 137 4.1 Design of Boost Converter 138 4.2 DC Motor Drive Circuit 139 5 Results and Analysis 144 5.1 Comparative Analysis of MPPT Algorithms 144 5.2 DC Motor Conveyor and Converter System 146 6 Conclusion 149 Acknowledgements 149 References 149 Differential Evolution Based Significant Data Region Identification on Large Storage Drives 151 1 Introduction 152 1.1 Motivation and Focus 153 2 Background and Related Work 154 2.1 Digital Forensics 156 2.2 Computational Intelligence Paradigm 159 3 Implementation of Differential Evolution Based Significant Data Region Identification 161 4 Experimental Setup 164 4.1 Experimental Results and Analysis 166 4.2 Case Study: Examination of Formatted Storage Drive 171 5 Discussion and Future Scope 172 6 Conclusion 174 References 175 9 A Fuzzy Based Power Switching Selection for Residential Application to Beat Peak Time Power Demand 176 Abstract 176 1 Introduction 177 2 Automatic Selection of Power Source During Peak Time 179 3 Fuzzy Logic in Automatic Selection of Power Source During Peak Time 182 4 Results and Analysis of the Design 185 5 Conclusion 187 References 187 10 Energy Saving Using Memorization: A Novel Energy Efficient and Fault Tolerant Cluster Tree Algorithm for WSN 189 Abstract 189 1 Introduction 190 1.1 Energy Concerns of Older Communication Protocols in WSNs 190 1.2 Advantage of Cluster-Tree Protocols 191 2 Related Work 192 2.1 TREEPSI [12] 192 2.2 TBDCS [13] 193 2.3 GTC 193 2.4 OCTBR [20] 194 2.5 CTDGA [10] 194 2.6 CIDT [21] 195 2.7 VELCT [22] 196 2.8 CTDD [23] 196 2.9 CTRP [24] 197 3 Comparative Chart of Topology Building Approaches in WSN 197 4 Research Gap 197 5 Proposed Solution ESUM 200 5.1 ESUM Algorithm 201 5.2 Energy Efficiency 202 5.3 Fault Tolerance 202 5.4 Comparison of Energy Requirements After Initial Topology Setup 203 5.5 Comparison of Energy Requirements for Re-election 206 5.6 Node Localization in WSN 206 5.7 Using Fuzzy Inference System to Overcome Problems of Crisp Programming 209 6 Advantages of ESUM 210 7 Conclusion 211 Appendix A 212 Appendix B: Activity Diagrams 213 References 215 Analyzing Slavic Textual Sentiment Using Deep Convolutional Neural Networks 217 1 Motivation 218 2 Introduction 218 2.1 Slavic Languages Analysis Characteristics 219 2.2 Data and Privacy 221 3 Textual Sentiment Analysis: Methods 221 3.1 Basic Linear, Binary, ReLU, Tanh and Sigmoid Model 223 3.2 Basic Neural Network Deployment Setting 224 3.3 Backpropagation 225 3.4 Convolutional Neural Networks 227 3.5 Sentiment Analysis 227 3.6 Problem with Deep Structures 228 4 Textual Sentiment Analysis in Slavic Languages: The Model 228 4.1 Model Results and Comparison with Dataset Benchmarks 229 4.2 Model Sustainability 230 4.3 Data Sustainability and Big Data 230 4.4 Privacy Issues 231 5 Conclusion 232 References 233 12 Intelligent Decision Support System for an Integrated Pest Management in Apple Orchard 235 Abstract 235 1 Introduction 236 2 Background 237 3 Integrated Pest Management (IPM) 238 4 Computational Intelligence 238 4.1 Case Based Reasoning 239 4.2 CBR Techniques 241 5 Towards Computational Intelligent DSS for IPM 242 5.1 Computational Intelligence Techniques Used 243 5.2 Description of the Intelligent IPM DSS 244 6 Results and Discussion 247 7 Scope, Significance and Limitations of the Study 252 8 Conclusions 252 References 253 13 Analysis of Error Propagation in Safety Critical Software Systems: An Approach Based on UGF 256 Abstract 256 1 Introduction 256 2 Terminologies 257 3 Related Works 258 4 Influence of EP in Software Reliability Prediction 260 5 Universal Generating Function Technique for MS 260 6 Problem Statement 262 7 Proposed Approach 262 8 EP and Failure Analysis 262 8.1 Performance Distribution of a Module 264 8.2 Safety Metric SMEP 265 8.3 Module Definition in Terms of PDMOD and SMEP 266 8.4 Performance Distribution of Subsystem PDSS 266 9 Case Study 267 10 Conclusion and Future Work 268 References 269 14 A Framework for Analyzing Uncertainty in Data Using Computational Intelligence Techniques 271 Abstract 271 1 Introduction 271 1.1 Motivation 272 1.2 Objectives of the Proposed Work 272 1.3 Computational Models for Prediction 273 2 Uncertainty in Data 273 2.1 Classical Set Theory Versus Fuzzy Set Theory Versus Rough Set Theory 274 2.2 Combining Fuzzy and Rough Set Theories 276 2.3 Neural Networks 277 2.4 Genetic Algorithms 279 2.5 Hybrid Systems 279 3 Classification Modeling Using CI Techniques 281 3.1 Proposed Algorithm for Rule Generation 282 3.2 Multi-layer Perceptron for Classification 284 3.3 Genetic-Fuzzy Modeling (GFM) 284 3.4 Rule Classifier Using Fuzzy Ant Colony Optimization 286 3.5 Fuzzy Discrete Particle Swarm Optimization Classifier for Rule Classification 287 4 Experimental Evaluation of the Classification Model 288 4.1 Evaluation Metrics 288 4.2 Description of the Dataset Used 288 4.3 Experimental Analysis 289 5 Conclusions and Future Research Directions 290 References 290 Index 294 Front Matter....Pages i-xvi Intelligent Decision Support Systems for Sustainable Computing....Pages 1-6 A Genetic Algorithm Based Efficient Static Load Distribution Strategy for Handling Large-Scale Workloads on Sustainable Computing Systems....Pages 7-31 Efficiency in Energy Decision Support Systems Using Soft Computing Techniques....Pages 33-52 Computational Intelligence Based Heuristic Approach for Maximizing Energy Efficiency in Internet of Things....Pages 53-76 Distributed Algorithm with Inherent Intelligence for Multi-cloud Resource Provisioning....Pages 77-99 Parameter Optimization Methods Based on Computational Intelligence Techniques in Context of Sustainable Computing....Pages 101-113 The Maximum Power Point Tracking Using Fuzzy Logic Algorithm for DC Motor Based Conveyor System....Pages 115-138 Differential Evolution Based Significant Data Region Identification on Large Storage Drives....Pages 139-163 A Fuzzy Based Power Switching Selection for Residential Application to Beat Peak Time Power Demand....Pages 165-177 Energy Saving Using Memorization: A Novel Energy Efficient and Fault Tolerant Cluster Tree Algorithm for WSN....Pages 179-206 Analyzing Slavic Textual Sentiment Using Deep Convolutional Neural Networks....Pages 207-224 Intelligent Decision Support System for an Integrated Pest Management in Apple Orchard....Pages 225-245 Analysis of Error Propagation in Safety Critical Software Systems: An Approach Based on UGF....Pages 247-261 A Framework for Analyzing Uncertainty in Data Using Computational Intelligence Techniques....Pages 263-285 Back Matter....Pages 287-289