Machine Learning in Cognitive IoT

go to site Cover Half Title Title Page Copyright Page Dedication Table of Contents Preface Acknowledgements List of Figures List of Tables 1 Internet of Things 1.1 IoT History 1.2 IoT Architecture 1.3 IoT Elements 1.3.1 Wireless Sensor Networks (WSN) 1.3.2 Radio Frequency Identification (RFID) 1.3.2.1 RFID Applications 1.3.3 Data Storage 1.3.4 Challenging Issues 1.4 Data Analytics 1.4.1 IoT Data Sources 1.4.1.1 Industrial Data 1.4.1.2 Business Applications 1.4.1.3 Sensors and Devices 1.4.1.4 Smartphones 1.4.2 Data Processing 1.4.2.1 Data Acquisition 1.4.2.2 Data Transport 1.4.2.3 Data Preprocessing 1.4.3 IoT Technologies 1.4.3.1 Wireless Communication Technologies 1.4.3.2 Data Analysis Schemes 1.4.3.3 Machine Learning 1.4.4 Optimization Techniques 1.4.4.1 Edge Computing Technology 1.4.4.2 Refining the Results 1.5 Steps of Data Preprocessing 1.5.1 Data Formatting 1.5.2 Data Cleaning 1.5.2.1 Duplicate Observations 1.5.2.2 Irrelevant Observations 1.5.2.3 Handling Missing Data 1.5.2.4 Detecting Outliers 1.5.3 Data Reduction Schemes 1.5.3.1 Feature Extraction 1.5.3.2 Feature Selection 1.6 IoT Protocols 1.6.1 Infrastructure Layer (Network/Transport Layer) 1.6.2 Data Protocols 1.6.3 Physical Layer 1.6.4 LPWAN: Low Power Wide Area Network 1.7 IoT Applications 1.7.1 Logistics and Transportation 1.7.2 Home and Workplace 1.7.3 Personal and Social 1.7.4 Health Domain 1.8 Book Outline 1.9 Target Audience 1.10 Summary and What’s Next? 1.11 Exercises 2 Cognitive Internet of Things 2.1 Cognitive Devices 2.2 Cognitive in IoT 2.3 CIoT Background 2.4 CIoT Elements 2.4.1 Sensors 2.4.2 Machine Learning 2.4.3 Cloud Storage 2.5 How Do Cognitive Devices Act as Human Assistants? 2.6 Machine-to-machine Interfaces 2.6.1 Language 2.6.2 Interpersonal Relationship 2.7 Man-to-machine Communication 2.8 Machine-to-web Communication(M2W) 2.9 CIoT Applications 2.9.1 Cognitive Living 2.9.2 Cognitive Cities 2.9.3 Cognitive Health 2.9.4 Auto-casting and Auto-reacting Cognition Systems 2.10 Summary and What’s Next? 2.11 Exercises 3 Data Mining in IoT 3.1 Search Engines as a Medium 3.2 Data Creation and Retrieval Scheme 3.3 Data Mining 3.3.1 Data Mining Functions 3.3.1.1 Classification 3.3.2 Relation of Data Science with Machine Learning 3.4 Data Mining in IoT 3.5 Machine Learning in IoT 3.6 Summary and What’s Next? 3.7 Exercises 4 Machine Learning Techniques 4.1 Tools to Implement Machine Learning 4.1.1 Python 4.1.2 R 4.1.3 Matlab 4.1.4 Weka 4.2 Experiments 4.2.1 Dataset 4.3 Supervised Learning 4.3.1 Unsupervised Learning 4.4 Classification 4.5 Regression 4.6 Clustering 4.7 Summary and What’s Next? 4.8 Exercises 5 R Programming 5.1 Introduction 5.1.1 Basis of the R programming 5.1.2 Installing R 5.1.3 Working in R 5.2 Basic Commands 5.2.1 Assignment 5.2.2 Comments 5.3 Data Types 5.3.1 Basic Data Types 5.3.1.1 Numbers 5.3.1.2 String 5.3.2 Structural Data Type 5.3.2.1 Matrices 5.3.2.2 Arrays 5.3.2.3 Data Frames 5.3.2.4 Lists 5.3.2.5 Factors 5.3.2.6 Vectors 5.3.3 Operators 5.3.3.1 Arithmetic Operators 5.3.3.2 Relational Operators 5.3.3.3 Logical Operators 5.3.4 Graphics 5.3.4.1 Histograms 5.3.4.2 Scatter Diagrams 5.3.4.3 Pie Chart 5.3.5 Basic Statistics 5.3.5.1 Descriptive Statistics 5.3.5.2 Correlation Testing 5.3.6 Packages 5.3.7 Input Parameters Formats for R 5.4 Summary and What’s Next? 5.5 Exercises 6 Machine Learning Paradigms 6.1 Introduction 6.2 Generalizing Input 6.3 Generalizing Output 6.3.1 Decision Tree 6.4 Classification Rules 6.5 Numeric Prediction 6.6 Instance-based Learning 6.6.1 Distance Metric 6.7 Summary and What’s Next 6.8 Exercises 7 Different Machine Learning Models 7.1 Linear Method for Regression 7.2 Linear Method for Classification 7.3 Kernel Smoothing Models 7.4 Back Propagation 7.4.1 Radial Basis Function Networks 7.5 Neural Network 7.5.1 The Perceptron 7.6 Bayesian Methods 7.6.1 Bayesian Statistics 7.6.2 Bayesian Inference 7.7 Summary and What’s Next 7.8 Exercises 8 Data Processing 8.1 Input Preparation 8.2 Data Preprocessing 8.3 Data Cleaning 8.3.1 The Condensed Nearest Neighbor Rule 8.3.2 Tomek 8.3.3 One-sided Selection 8.3.4 SMOTE 8.3.5 ADASYN Algorithm 8.3.6 SOTU 8.4 Summary and What’s Next? 8.5 Exercises 9 Feature Engineering and Optimization 9.1 Feature Reduction 9.1.1 Principal Component Analysis 9.2 Feature Selection 9.2.1 Feature Importance 9.2.1.1 Chi-squared Filter 9.2.1.2 Consistency-based Filter 9.2.1.3 Correlation Filter 9.2.1.4 Entropy-based Filter 9.2.1.5 OneR Algorithm 9.2.1.6 RandomForest Filter 9.2.1.7 RReliefF Filter 9.2.2 Recursive Feature Elimination 9.3 Machine Learning Models 9.3.1 Experiments 9.4 Bagging and Boosting Techniques 9.4.1 Bagging 9.4.2 Boosting 9.5 Ensemble Approach 9.6 Summary and What’s Next 9.7 Exercises 10 Evaluation and Validation of Results 10.1 Confusion Matrix 10.2 Correlation 10.2.1 Covariance 10.2.2 Pearson’s Correlation 10.2.3 Spearman’s Correlation 10.2.4 Matthews’ Correlation Coefficient (MCC) 10.3 Coefficient of Determinant: R2 10.4 Accuracy (ACC) 10.5 ROC and AUC 10.6 Error in Regression 10.6.1 Root Mean Squared Error (RMSE) 10.6.2 Mean Absolute Error (MAE) 10.6.3 Relative Squared Error (RSE) 10.6.4 Relative Absolute Error (RAE) 10.7 Measuring Rates 10.7.1 Sensitivity, Recall, Hit Rate, or True Positive Rate (TPR) 10.7.2 Specificity, Selectivity, or True Negative Rate (TNR) 10.7.3 Precision, Positive Predictive Value (PPV) 10.7.4 Recall, Sensitivity, Hit Rate, or True Positive Rate (TPR) 10.7.5 Fallout, False Positive Rate (FPR) 10.7.6 Miss Rate or False Negative Rate (FNR) 10.7.7 False Discovery Rate (FDR) 10.7.8 False Omission Rate (FOR) 10.8 F Measure 10.9 Summary and What’s Next? 10.10 Exercises 11 Solutions 11.1 Chapter 1 11.2 Chapter 2 11.3 Chapter 3 11.4 Chapter 4 11.5 Chapter 5 11.6 Chapter 6 11.7 Chapter 7 11.8 Chapter 8 11.9 Chapter 9 11.10 Chapter 10 12 Dataset Bibliography Index