Big Data, IoT, and Machine Learning: Tools and Applications (Internet of Everything (IoE))

✍ Scribed by Rashmi Agrawal (editor), Marcin Paprzycki (editor), Neha Gupta (editor)

• Publisher: CRC Press
• Year: 2020
• Tongue: English
• Leaves: 339
• Series: Internet of Everything (IoE)
• Edition: 1
• Category: Library

✦ Synopsis

The idea behind this book is to simplify the journey of aspiring readers and researchers to understand Big Data, IoT and Machine Learning. It also includes various real-time/offline applications and case studies in the fields of engineering, computer science, information security and cloud computing using modern tools.

This book consists of two sections: Section I contains the topics related to Applications of Machine Learning, and Section II addresses issues about Big Data, the Cloud and the Internet of Things. This brings all the related technologies into a single source so that undergraduate and postgraduate students, researchers, academicians and people in industry can easily understand them.

Features

• Addresses the complete data science technologies workflow
• Explores basic and high-level concepts and services as a manual for those in the industry and at the same time can help beginners to understand both basic and advanced aspects of machine learning
• Covers data processing and security solutions in IoT and Big Data applications
• Offers adaptive, robust, scalable and reliable applications to develop solutions for day-to-day problems
• Presents security issues and data migration techniques of NoSQL databases

✦ Table of Contents

Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Preface
Acknowledgement
Editors
Contributors
Section I Applications of Machine Learning
Chapter 1 Machine Learning Classifiers
1.1 Introduction
1.2 Machine Learning Overview
1.2.1 Steps in Machine Learning
1.2.2 Performance Measures for Machine Learning Algorithms
1.2.2.1 Confusion Matrix
1.3 Machine Learning Approaches
1.4 Types of Machine Learning
1.4.1 Supervised Learning
1.4.2 Unsupervised Learning
1.4.3 Semi-Supervised Learning
1.4.4 Reinforcement Learning
1.5 A Taste of Classification
1.5.1 Binary Classification
1.5.2 Multiclass Classification
1.5.3 Multilabel Classification
1.5.4 Linear Classification
1.5.5 Non-Linear Classification
1.6 Machine Learning Classifiers
1.6.1 Python for Machine Learning Classification
1.6.2 Decision Tree
1.6.2.1 Building a Decision Tree
1.6.2.2 Induction
1.6.2.3 Best Attribute Selection
1.6.2.4 Pruning
1.6.3 Random Forests
1.6.3.1 Evaluating Random Forest
1.6.3.2 Tuning Parameters in Random Forest
1.6.3.3 Splitting Rule
1.6.4 Support Vector Machine
1.6.5 Neural Networks
1.6.5.1 Back Propagation Algorithm
1.6.6 Logistic Regression
1.6.7 k-Nearest Neighbor
1.6.7.1 The k-NN Algorithm
1.7 Model Selection and Validation
1.7.1 Hyperparameter Tuning and Model Selection
1.7.2 Bias, Variance and Model Selection
1.7.3 Model Validation
Conclusion
References
Chapter 2 Dimension Reduction Techniques
2.1 Dimension Reduction
2.2 Dimension Reduction Techniques
2.2.1 Feature Selection
2.2.2 Feature Extraction
2.3 Linear Dimension Reduction Techniques
2.3.1 Principal Component Analysis
2.3.2 Singular Value Decomposition
2.3.3 Latent Discriminant Analysis
2.3.4 Independent Component Analysis
2.3.5 Projection Pursuits
2.3.6 Latent Semantic Analysis
2.3.7 Locality Preserving Projection
2.4 Nonlinear Dimension Reduction Techniques
2.4.1 Kernel Principal Component Analysis
2.4.2 Isomap
2.4.3 Locally Linear Embedding
2.4.4 Self Organising Map
2.4.5 Learning Vector Quantisation
2.4.6 t-Stochastic Neighbor Embedding
2.5 Conclusion and Future Directions
References
Chapter 3 Reviews Analysis of Apple Store Applications Using Supervised Machine Learning
3.1 Introduction
3.2 Literature Review
3.2.1 Machine Learning Algorithms
3.2.2 Feature Extraction Algorithms
3.3 Proposed Methodology
3.3.1 Data Collection
3.3.2 Feature Extraction
3.3.3 Data Analysis and Sentiment Analysis
Text Processing
3.3.4 Text Normalisation
3.4 Feature Extraction Algorithm
3.4.1 CountVectorizer
3.4.2 TfidfVectorizer (TF–IDF)
3.5 Supervised ML Classification
3.6 Experiment Design
3.7 Experimental Results and Analysis
3.8 Recommendation and Future Work
3.9 Conclusion
References
Chapter 4 Machine Learning for Biomedical and Health Informatics
4.1 Introduction
4.2 Overview of Machine Learning Applications
4.3 Impact of Machine Learning in Healthcare
4.4 Recent Trends in Machine Learning in the Biomedical Field
4.6 Supervised Learning Methods
4.7 Unsupervised Learning Methods
4.8 Reinforcement Learning (RL)
4.9 Semi-Supervised Learning
4.9.1 K-Nearest Neighbor (KNN)
4.9.2 Naive Bayes (NB)
4.9.3 Decision Trees (DT)
4.9.4 Support Vector Machine (SVM)
4.9.5 Artificial Neural Network (ANN)
4.10 Deep Learning (DL)
4.10.1 Recurrent Neural Network (RNN)
4.10.2 Convolutional Neural Network (CNN)
4.10.3 Deep Learning in Healthcare
4.11 Existing Works on ML for Biomedical and Health Informatics
4.12 Conclusion and Future Issues
References
Chapter 5 Meta-Heuristic Algorithms: A Concentration on the Applications in Text Mining
5.1 Introduction
5.2 Literature Review of Meta-Heuristic Algorithms
5.2.1 Genetic Algorithms (GA)
5.2.2 Ant Colony Optimisation (ACO)
5.2.3 Ant Lion Optimiser (ALO)
5.2.4 Bat Algorithm (BA)
5.2.5 Cat Swarm Optimisation Algorithm (CSO)
5.2.6 Crow Search Algorithm (CSA)
5.2.7 Cuckoo Optimisation Algorithm (COA)
5.2.8 Bee Colony Optimisation (BCO)
5.2.9 Particle Swarm Optimisation (PSO)
5.2.10 Firefly Algorithm (FA)
5.2.11 Tabu Search Algorithm (TS)
5.3 Proposed Model for Application of Meta-Heuristic in Text Mining
5.4 Future Research
5.5 Conclusion
References
Chapter 6 Optimizing Text Data in Deep Learning: An Experimental Approach
6.1 Introduction
6.2 Existing Structure of Deep Learning
6.2.1 Neural Networks
6.3 Problems in Existing Definition
6.4 Research Trust
6.5 Text Classification
6.5.1 Steps of Text Classification
6.5.2 Developing a GUI-Based Deep Learning Application to Perform Text Classification on Reuters Dataset
6.6 Experimentation
6.6.1 Code
6.7 Conclusion and Future Scope
References
Section II Big Data, Cloud and Internet of Things
Chapter 7 Latest Data and Analytics Technology Trends That Will Change Business Perspectives
7.1 Introduction
7.2 Strategic Planning Assumptions and Analysis
7.3 Driving Factors for Latest Data and Analytics Technology Trends
7.3.1 Trend 1: Augmented Analytics
7.3.1.1 What Does It Enable?
7.3.1.2 Use Cases
7.3.1.3 Recommendations
7.3.2 Trend 2: Augmented Data Management
7.3.2.1 What Does It Enable?
7.3.2.2 How Does This Impact Your Organisation and Skills?
7.3.2.3 Use Cases
7.3.2.4 Recommendations
7.3.3 Trend 3: NLP and Conversational Analytics
7.3.3.1 What Does It Enable?
7.3.3.2 How Does This Impact Your Organisation and Skills?
7.3.3.3 Use Cases
7.3.3.4 Recommendations
7.3.4 Trend 4: Graph Analytics
7.3.4.1 What Does It Enable?
7.3.4.2 How Does This Impact Your Organisation and Skills?
7.3.4.3 Use Cases
7.3.5 Trend 5: Commercial AI/ML Will Dominate the Market over Open Source
7.3.5.1 What Does It Enable?
7.3.5.2 How Does This Impact Your Organisation and Skills?
7.3.5.3 Use Cases
7.3.5.4 Recommendations
7.3.6 Trend 6: Data Fabric
7.3.6.1 What Does It Enable?
7.3.6.2 How Does This Impact Your Organisation and Skills?
7.3.6.3 Use Cases
7.3.6.4 Recommendations
7.3.7 Trend 7: Explainable AI
7.3.7.1 What Does It Enable?
7.3.7.2 How Does This Impact Your Organisation and Skills?
7.3.7.3 Use Cases
7.3.7.4 Recommendations
7.3.8 Trend 8: Blockchain in Data and Analytics
7.3.8.1 What Does It Enable?
7.3.8.2 How Does This Impact Your Organisation and Skills?
7.3.8.3 Use Cases
7.3.8.4 Recommendations
7.3.9 Trend 9: Continuous Intelligence
7.3.9.1 What Does It Enable?
7.3.9.2 How Does This Impact Your Organisation and Skills?
7.3.9.3 Use Cases
7.3.9.4 Recommendations
7.3.10 Trend 10: Persistent Memory Servers
7.3.10.1 What Does It Enable?
7.3.10.2 How Does This Impact Your Organisation and Skills?
7.3.10.3 Use Cases
7.3.10.4 Recommendations
References
Chapter 8 A Proposal Based on Discrete Events for Improvement of the Transmission Channels in Cloud Environments and Big Data
8.1 Introduction
8.2 Big Data
8.3 Cloud Computing
8.4 Big Data and Cloud Computing
8.5 Discrete Event, Communication Channel and Modulation
8.6 Methodology
8.7 Results and Discussion
8.8 Future Research Directions
8.9 Conclusion
References
Chapter 9 Heterogeneous Data Fusion for Healthcare Monitoring: A Survey
9.1 Introduction
9.2 Sensor Data Fusion
9.2.1 Sensor Data Fusion in the Healthcare Environment
9.3 Healthcare Data Fusion: Opportunities and Challenges
9.3.1 Healthcare Data Fusion: Opportunities
9.3.2 Healthcare Data Fusion: Challenges
9.4 Evaluation Framework
9.4.1 Middleware Architecture Type
9.4.2 Context Awareness
9.4.3 Semantic Interaction
9.4.4 Dynamic Configuration
9.4.5 Fusion Complexity
9.4.6 Actuation Management
9.4.7 Data Processing Type
9.4.8 Cross Domain Portability
9.4.9 Implementation
9.4.10 Performance Evaluation
9.4.11 Data Security and Privacy
9.5 Application of Data Fusion for Health Monitoring
9.6 Conclusion
References
Chapter 10 Discriminative and Generative Model Learning for Video Object Tracking
10.1 Introduction: Artificial Intelligence and Computer Vision
10.2 Computer Vision
10.3 Introduction to Video Object Tracking
10.4 Appearance Model of the Target
10.4.1 Construction of Generative Appearance Model
10.4.2 Generative Appearance Model
10.5 Motion Model
10.6 Proposed Method of Online Parameter Learning
10.7 Experimental Results
10.8 Conclusion
References
Chapter 11 Feature, Technology, Application, and Challenges of Internet of Things
11.1 Introduction
11.2 About the Web of Things
11.3 IoT as Tool for Change in Technology
11.4 Characteristics of IoT
11.4.1 Heterogeneity
11.4.2 Interconnectivity
11.4.3 Dynamic Changes
11.4.4 Enormous Scale
11.4.5 Safety
11.4.6 Connectivity
11.5 Applications of IoT
11.5.1 Smarter Cities
11.5.2 Smarter Home
11.5.3 Smart Energy
11.5.4 Smart Health
11.5.5 Environmental Observation (Smart Appliances)
11.5.6 Smart Vesture and Good Accessories (Wearable)
11.5.7 Hobbyists
11.6 Challenges
11.6.1 Scalability
11.6.2 Self-Organizing
11.6.3 Data Volumes
11.6.4 Data Interpretation
11.6.5 Interoperability
11.6.6 Automatic Discovery
11.6.7 Software Complexity
11.6.8 Security and Privacy
11.6.9 Wireless Communications
11.7 The Problem of Overlays
11.7.1 Redundant Overlay Networks
11.7.2 Management Complexity
11.7.3 Precious Physical Space
11.8 Rise of Converged APs
11.9 Addressing Convergence Challenges
11.9.1 Radiofrequency (RF) Interference
11.9.2 Packet Coordination
11.9.3 Antenna Design
11.10 Future Technologies of IoT
11.10.1 Cloud Computing
11.10.2 Shared Computing
11.10.3 Cloud Computing
11.10.4 Wireless Fidelity (Wi-Fi)
11.10.5 Bluetooth
11.10.6 ZigBee
11.11 Cloud Computing in IoT
11.11.1 Remote Process Power
11.11.2 Lowers the Entry Bar for Suppliers Who Lack the Infrastructure
11.11.3 Analytics and Observation
11.11.4 User Security and Privacy
11.12 Challenges in Integration of Cloud Computing and IoT
11.12.1 No Uniformity
11.12.2 Performance
11.12.3 Dependableness
11.12.4 Massive Scale
11.12.5 Big Data
11.13 Conclusion
References
Chapter 12 Analytical Approach to Sustainable Smart City Using IoT and Machine Learning
12.1 Introduction to Smart City
12.2 Background
12.3 Smart City Architecture
12.3.1 Sensing Layer
12.3.2 Transmission Layer
12.3.3 Data Management Layer
12.3.4 Application Layer
12.4 Major Smartest Cities in the World
12.4.1 Reykjavik
12.4.2 Tokyo
12.4.3 Paris
12.4.4 London
12.4.5 New York City
12.5 Role of Fog Computing in Smart City
12.6 Analytical Approach to Sustainability in the Smart City
12.7 Enabling Technologies for Sustainability
12.7.1 IoT
12.7.2 Machine Learning
12.7.3 Big Data
12.8 Proposed Model for the Analytical Framework of a Sustainable Smart City
12.9 Conclusion
References
Chapter 13 Traffic Flow Prediction with Convolutional Neural Network Accelerated by Spark Distributed Cluster
13.1 Introduction
13.2 Background and Related Studies
13.2.1 Machine Learning
13.2.2 Deep Learning
13.2.3 CNN
13.2.4 Spark Cluster and Distributed Environment Acceleration
13.3 Existing Machine Learning and Deep Learning Methods
13.3.1 Decision Tree
13.3.2 Random Forest
13.3.3 SVM
13.3.4 KNN
13.3.5 CNN
13.3.6 Comparison of Performance
13.4 Proposed Method: CNN with Spark
13.4.1 Workflow
13.4.2 CNN Model Design and Modification
13.4.2.1 Learning Rate
13.4.2.2 Activation Function
13.4.2.3 Pooling Layer
13.4.2.4 Final CNN Model
13.4.3 Spark Cluster Configuration
13.4.3.1 Four Modes
13.4.3.2 Memory Layout
13.5 Performance Evaluation
13.5.1 Experiment Setup
13.5.1.1 Dataset
13.5.1.2 Profiling Tool
13.5.1.3 Four Measures
13.5.2 Results and Analysis
13.5.2.1 CNN Model Optimisation
13.5.2.2 Spark Cluster Tuning
13.5.3 Summary of Performance Evaluation
13.6 Conclusion and Future Directions
References
Index