Big Data Analytics for Smart Urban Systems (Urban Sustainability)

Big Data Analytics for Smart Urban Systems aims to introduce Big data solutions for urban sustainability smart applications, particularly for smart urban systems. It focuses on intelligent big data which takes the benefits of machine learning to analyse large and rapidly changing datasets in smart urban systems. The state-of-the-art Big data analytics applications are presented and discussed to highlight the feasibility of big data and machine learning solutions to enhance smart urban systems, smart operations, urban management, and urban governance. The key benefits of this book are, (1) to introduce the principles of machine learning-enabled big data analysis in smart urban systems, (2) to present the state-of-the-art data analysis solutions in smart management and operations, and (3) to understand the principles of big data analytics for smart cities and communities. Endorsements ‘Over the many years of collaboration between academia and industry, we noticed the common language is ‘big data’; with that, we have developed novel ideas to bridge the gaps and help promote innovation, technologies, and science’.- Tian Tang, Independent Researcher, China ‘Big Data Analytics is a fascinating research area, particularly for cities and city transformations. This book is valuable to those who think vigorously and aim to act ahead’.- Li Xie, Independent Researcher, China ‘For urban critiques, knowledge trains aspiring opportunities toward outstanding manifestations. Smartness has evolved or/ advanced rambunctious & embracing realities along (with) novel directions and nurturing integrated city knowledge’.- Aaron Golden, SELECT Consultants, UK Preface 7 Acknowledgements 9 About This Book 10 Praise for Big Data Analytics for Smart Urban Systems 11 Contents 12 About the Authors 15 1 Big Data Analytics: An Introduction to Their Applications for Smart Urban Systems 16 1.1 The Emergence of Big Data Analytics 16 1.2 The Aim and Objectives of the Book 17 1.3 The Structure of Two Volumes on Big Data Analytics 18 1.4 A Summary 21 Box 1.1 Examples of ‘Smart Cities’ reports and documents 25 Box 1.2 Examples of ‘Smart Cities’ reports and documents 26 Box 1.3 Examples of ‘Smart Cities’ reports and documents 27 Box 1.4 Examples of ‘Smart Cities’ reports and documents 28 Box 1.5 Examples of ‘Smart Cities’ reports and documents 29 Box 1.6 Examples of ‘Smart Cities’ reports and documents 30 Box 1.7 Examples of ‘Smart Cities’ reports and documents 31 Box 1.8 Examples of ‘Smart Cities’ reports and documents 32 Box 1.9 Examples of ‘Smart Cities’ reports and documents 33 Box 1.10 Examples of ‘Smart Cities’ reports and documents 34 References 34 2 Stock Market Prediction During COVID-19 Pandemic: A Time-Series Big Data Analysis Method 37 2.1 Introduction 37 2.2 Literature Review 38 2.2.1 Big Data Analytics in Stock Markets 39 2.3 Methodology 40 2.3.1 Data Preprocessing 41 2.3.2 Pattern Retrieval Using DTW 41 2.3.3 Feature Selection 42 2.3.4 Predicted Stock Data Generation Using LSTM 43 2.4 Result Analysis and Discussion 44 2.4.1 Data Preprocessing 44 2.4.2 Estimation of Close Price and COVID-19 Data 45 2.4.3 Pattern Selection 46 2.4.4 Feature Selection Result with Analysis 47 2.4.5 Result for LSTM Price Prediction 48 2.4.6 Predicted Price and Covid-19 Data Factors 49 2.5 Conclusion 50 References 52 3 A Big Data Solution to Predict Cryptocurrency Market Trends: A Time-Series Machine Learning Approach 54 3.1 Introduction 54 3.2 Literature Review 55 3.2.1 Cryptocurrency Pattern Recognition and Clustering 55 3.2.2 Bitcoin Price Prediction 56 3.3 Methodology 56 3.3.1 Dataset Selection and Pre-processing 57 3.3.2 Data Pattern Recognition via Clustering 58 3.3.3 Predictive Analysis 60 3.4 Result and Discussion 62 3.4.1 Trend Prediction 62 3.5 Conclusion 65 References 65 4 Big Data Analytics for Credit Risk Prediction: Machine Learning Techniques and Data Processing Approaches 68 4.1 Introduction 68 4.2 Literature Review 69 4.3 Methodology 70 4.3.1 Dataset and Data Pre-processing 70 4.3.2 Machine Learning Models 76 4.4 Result and Discussion 76 4.5 Conclusion 77 References 78 5 Worldwide Mobility Trends and the COVID-19 Pandemic: A Federated Regression Analysis During the pandemic’s Early Stage 80 5.1 Introduction 80 5.2 Literature Review on Existing Research Studies 81 5.2.1 Influence Factors 81 5.2.2 Pharmacological and Non-pharmacological Interventions 81 5.2.3 Social Distance Policy 82 5.2.4 Reflection of H1N1 82 5.2.5 Cultural Susceptibility and Policy 83 5.2.6 Voluntary Mechanisms 84 5.3 Methodology 85 5.3.1 Data Sources 85 5.3.2 Statistical Analysis 85 5.3.3 Data Analysis 86 5.3.4 Correlation Matrix 86 5.3.5 Regression 86 5.4 Results and Discussion 86 5.4.1 Correlation 86 5.4.2 Regression Results 89 5.5 Conclusions 89 References 91 6 Adaptive Feature Selection for Google App Rating in Smart Urban Management: A Big Data Analysis Approach 93 6.1 Introduction 93 6.2 Literature Review 94 6.2.1 Traditional Dimension Reduction Techniques 94 6.2.2 Random Forest 96 6.2.3 Data Pre-processing 98 6.3 Methodology 99 6.4 Results and Discussions 101 6.4.1 Overall Comparison 101 6.4.2 Discussion on Random Forest 102 6.4.3 Discussion on Linear Discriminant Analysis 104 6.5 Conclusions 105 References 106 7 Improve the Daily Societal Operations Using Credit Fraud Detection: A Big Data Classification Solution 109 7.1 Introduction: An Overview of Recent and Ongoing Research on Credit Fraud Detection 109 7.2 Literature Review Related to Big Data and Credit Fraud Detection 110 7.3 Methodology 112 7.3.1 Dataset Introduction 112 7.3.2 Data Preprocess and Feature Extraction 113 7.3.3 Model Description 114 7.3.4 Model Implementation 116 7.4 Results and Analysis 117 7.5 Conclusions 119 References 120 8 Moving Forward with Big Data Analytics and Smartness 123 8.1 A Brief Reflection on Big Data Analytics and Smart Urban Systems 123 8.2 Methodological Contributions of the Book 132 8.3 Concluding Remarks: A Summary of Lessons Learnt for Future Research 133 References 136 Index 139