Machine Learning and Big Data: Concepts, Algorithms, Tools and Applications

Cover
Title Page
Copyright Page
Contents
Preface
Section 1: Theoretical Fundamentals
	Chapter 1 Mathematical Foundation
		1.1 Concept of Linear Algebra
			1.1.1 Introduction
			1.1.2 Vector Spaces
			1.1.3 Linear Combination
			1.1.4 Linearly Dependent and Independent Vectors
			1.1.5 Linear Span, Basis and Subspace
			1.1.6 Linear Transformation (or Linear Map)
			1.1.7 Matrix Representation of Linear Transformation
				1.1.7.1 Transformation Matrix
			1.1.8 Range and Null Space of Linear Transformation
			1.1.9 Invertible Linear Transformation
		1.2 Eigenvalues, Eigenvectors, and Eigendecomposition of a Matrix
			1.2.1 Characteristics Polynomial
				1.2.1.1 Some Results on Eigenvalue
			1.2.2 Eigendecomposition [11]
		1.3 Introduction to Calculus
			1.3.1 Function
			1.3.2 Limits of Functions
				1.3.2.1 Some Properties of Limits
				1.3.2.2 1nfinite Limits
				1.3.2.3 Limits at Infinity
			1.3.3 Continuous Functions and Discontinuous Functions
				1.3.3.1 Discontinuous Functions
				1.3.3.2 Properties of Continuous Function
			1.3.4 Differentiation
		References
	Chapter 2 Theory of Probability
		2.1 Introduction
			2.1.1 Definition
				2.1.1.1 Statistical Definition of Probability
				2.1.1.2 Mathematical Definition of Probability
			2.1.2 Some Basic Terms of Probability
				2.1.2.1 Trial and Event
				2.1.2.2 Exhaustive Events (Exhaustive Cases)
				2.1.2.3 Mutually Exclusive Events
				2.1.2.4 Equally Likely Events
				2.1.2.5 Certain Event or Sure Event
				2.1.2.6 Impossible Event or Null Event (.)
				2.1.2.7 Sample Space
				2.1.2.8 Permutation and Combination
				2.1.2.9 Examples
		2.2 Independence in Probability
			2.2.1 Independent Events
			2.2.2 Examples: Solve the Following Problems
		2.3 Conditional Probability
			2.3.1 Definition
			2.3.2 Mutually Independent Events
			2.3.3 Examples
		2.4 Cumulative Distribution Function
			2.4.1 Properties
			2.4.2 Example
		2.5 Baye’s Theorem
			2.5.1 Theorem
				2.5.1.1 Examples
		2.6 Multivariate Gaussian Function
			2.6.1 Definition
				2.6.1.1 Univariate Gaussian (i.e., One Variable Gaussian)
				2.6.1.2 Degenerate Univariate Gaussian
				2.6.1.3 Multivariate Gaussian
		References
	Chapter 3 Correlation and Regression
		3.1 Introduction
		3.2 Correlation
			3.2.1 Positive Correlation and Negative Correlation
			3.2.2 Simple Correlation and Multiple Correlation
			3.2.3 Partial Correlation and Total Correlation
			3.2.4 Correlation Coefficient
		3.3 Regression
			3.3.1 Linear Regression
			3.3.2 Logistic Regression
			3.3.3 Polynomial Regression
			3.3.4 Stepwise Regression
			3.3.5 Ridge Regression
			3.3.6 Lasso Regression
			3.3.7 Elastic Net Regression
		3.4 Conclusion
		References
Section 2: Big Data and Pattern Recognition
	Chapter 4 Data Preprocess
		4.1 Introduction
			4.1.1 Need of Data Preprocessing
			4.1.2 Main Tasks in Data Preprocessing
		4.2 Data Cleaning
			4.2.1 Missing Data
			4.2.2 Noisy Data
		4.3 Data Integration
			4.3.1 χ2 Correlation Test
			4.3.2 Correlation Coefficient Test
			4.3.3 Covariance Test
		4.4 Data Transformation
			4.4.1 Normalization
			4.4.2 Attribute Selection
			4.4.3 Discretization
			4.4.4 Concept Hierarchy Generation
		4.5 Data Reduction
			4.5.1 Data Cube Aggregation
			4.5.2 Attribute Subset Selection
			4.5.3 Numerosity Reduction
			4.5.4 Dimensionality Reduction
		4.6 Conclusion
		Acknowledgements
		References
	Chapter 5 Big Data
		5.1 Introduction
		5.2 Big Data Evaluation With Its Tools
		5.3 Architecture of Big Data
			5.3.1 Big Data Analytics Framework Workflow
		5.4 Issues and Challenges
			5.4.1 Volume
			5.4.2 Variety of Data
			5.4.3 Velocity
		5.5 Big Data Analytics Tools
		5.6 Big Data Use Cases
			5.6.1 Banking and Finance
			5.6.2 Fraud Detection
			5.6.3 Customer Division and Personalized Marketing
			5.6.4 Customer Support
			5.6.5 Risk Management
			5.6.6 Life Time Value Prediction
			5.6.7 Cyber Security Analytics
			5.6.8 Insurance Industry
			5.6.9 Health Care Sector
				5.6.9.1 Big Data Medical Decision Support
				5.6.9.2 Big Data–Based Disorder Management
				5.6.9.3 Big Data–Based Patient Monitoring and Control
				5.6.9.4 Big Data–Based Human Routine Analytics
			5.6.10 Internet of Things
			5.6.11 Weather Forecasting
		5.7 Where IoT Meets Big Data
			5.7.1 IoT Platform
			5.7.2 Sensors or Devices
			5.7.3 Device Aggregators
			5.7.4 IoT Gateway
			5.7.5 Big Data Platform and Tools
		5.8 Role of Machine Learning For Big Data and IoT
			5.8.1 Typical Machine Learning Use Cases
		5.9 Conclusion
		References
	Chapter 6 Pattern Recognition Concepts
		6.1 Classifier
			6.1.1 Introduction
			6.1.2 Explanation-Based Learning
			6.1.3 Isomorphism and Clique Method
			6.1.4 Context-Dependent Classification
			6.1.5 Summary
		6.2 Feature Processing
			6.2.1 Introduction
			6.2.2 Detection and Extracting Edge With Boundary Line
			6.2.3 Analyzing the Texture
			6.2.4 Feature Mapping in Consecutive Moving Frame
			6.2.5 Summary
		6.3 Clustering
			6.3.1 Introduction
			6.3.2 Types of Clustering Algorithms
				6.3.2.1 Dynamic Clustering Method
				6.3.2.2 Model-Based Clustering
			6.3.3 Application
			6.3.4 Summary
		6.4 Conclusion
		References
Section 3: Machine Learning: Algorithms & Applications
	Chapter 7 Machine Learning
		7.1 History and Purpose of Machine Learning
			7.1.1 History of Machine Learning
				7.1.1.1 What is Machine Learning?
				7.1.1.2 When the Machine Learning is Needed?
			7.1.2 Goals and Achievements in Machine Learning
			7.1.3 Applications of Machine Learning
				7.1.3.1 Practical Machine Learning Examples
			7.1.4 Relation to Other Fields
				7.1.4.1 Data Mining
				7.1.4.2 Artificial Intelligence
				7.1.4.3 Computational Statistics
				7.1.4.4 Probability
			7.1.5 Limitations of Machine Learning
		7.2 Concept of Well-Defined Learning Problem
			7.2.1 Concept Learning
				7.2.1.1 Concept Representation
				7.2.1.2 Instance Representation
				7.2.1.3 The Inductive Learning Hypothesis
			7.2.2 Concept Learning as Search
				7.2.2.1 Concept Generality
		7.3 General-to-Specific Ordering Over Hypotheses
			7.3.1 Basic Concepts: Hypothesis, Generality
			7.3.2 Structure of the Hypothesis Space
				7.3.2.1 Hypothesis Notations
				7.3.2.2 Hypothesis Evaluations
			7.3.3 Ordering on Hypotheses: General to Specific
				7.3.3.1 Most Specific Generalized
				7.3.3.2 Most General Specialized
				7.3.3.3 Generalization and Specialization Operators
			7.3.4 Hypothesis Space Search by Find-S Algorithm
				7.3.4.1 Properties of the Find-S Algorithm
				7.3.4.2 Limitations of the Find-S Algorithm
		7.4 Version Spaces and Candidate Elimination Algorithm
			7.4.1 Representing Version Spaces
				7.4.1.1 General Boundary
				7.4.1.2 Specific Boundary
			7.4.2 Version Space as Search Strategy
			7.4.3 The List-Eliminate Method
			7.4.4 The Candidate-Elimination Method
				7.4.4.1 Example
				7.4.4.2 Convergence of Candidate-Elimination Method
				7.4.4.3 Inductive Bias for Candidate-Elimination
		7.5 Concepts of Machine Learning Algorithm
			7.5.1 Types of Learning Algorithms
				7.5.1.1 Incremental vs. Batch Learning Algorithms
				7.5.1.2 Offline vs. Online Learning Algorithms
				7.5.1.3 Inductive vs. Deductive Learning Algorithms
			7.5.2 A Framework for Machine Learning Algorithms
				7.5.2.1 Training Data
				7.5.2.2 Target Function
				7.5.2.3 Construction Model
				7.5.2.4 Evaluation
			7.5.3 Types of Machine Learning Algorithms
				7.5.3.1 Supervised Learning
				7.5.3.2 Unsupervised Learning
				7.5.3.3 Semi-Supervised Learning
				7.5.3.4 Reinforcement Learning
				7.5.3.5 Deep Learning
			7.5.4 Types of Machine Learning Problems
				7.5.4.1 Classification
				7.5.4.2 Clustering
				7.5.4.3 Optimization
				7.5.4.4 Regression
		Conclusion
		References
	Chapter 8 Performance of Supervised Learning Algorithms on Multi-Variate Datasets
		8.1 Introduction
		8.2 Supervised Learning Algorithms
			8.2.1 Datasets and Experimental Setup
			8.2.2 Data Treatment/Preprocessing
		8.3 Classification
			8.3.1 Support Vector Machines (SVM)
			8.3.2 Naïve Bayes (NB) Algorithm
			8.3.3 Bayesian Network (BN)
			8.3.4 Hidden Markov Model (HMM)
			8.3.5 K-Nearest Neighbour (KNN)
			8.3.6 Training Time
		8.4 Neural Network
			8.4.1 Artificial Neural Networks Architecture
			8.4.2 Application Areas
			8.4.3 Artificial Neural Networks and Time Series
		8.5 Comparisons and Discussions
			8.5.1 Comparison of Classification Accuracy
			8.5.2 Forecasting Efficiency Comparison
			8.5.3 Recurrent Neural Network (RNN)
			8.5.4 Backpropagation Neural Network (BPNN)
			8.5.5 General Regression Neural Network
		8.6 Summary and Conclusion
		References
	Chapter 9 Unsupervised Learning
		9.1 Introduction
		9.2 Related Work
		9.3 Unsupervised Learning Algorithms
		9.4 Classification of Unsupervised Learning Algorithms
			9.4.1 Hierarchical Methods
			9.4.2 Partitioning Methods
			9.4.3 Density-Based Methods
			9.4.4 Grid-Based Methods
			9.4.5 Constraint-Based Clustering
		9.5 Unsupervised Learning Algorithms in ML
			9.5.1 Parametric Algorithms
			9.5.2 Non-Parametric Algorithms
			9.5.3 Dirichlet Process Mixture Model
			9.5.4 X-Means
		9.6 Summary and Conclusions
		References
	Chapter 10 Semi-Supervised Learning
		10.1 Introduction
			10.1.1 Semi-Supervised Learning
			10.1.2 Comparison With Other Paradigms
		10.2 Training Models
			10.2.1 Self-Training
			10.2.2 Co-Training
		10.3 Generative Models—Introduction
			10.3.1 Image Classification
			10.3.2 Text Categorization
			10.3.3 Speech Recognition
			10.3.4 Baum-Welch Algorithm
		10.4 S3VMs
		10.5 Graph-Based Algorithms
			10.5.1 Mincut
			10.5.2 Harmonic
			10.5.3 Manifold Regularization
		10.6 Multiview Learning
		10.7 Conclusion
		References
	Chapter 11 Reinforcement Learning
		11.1 Introduction: Reinforcement Learning
			11.1.1 Elements of Reinforcement Learning
		11.2 Model-Free RL
			11.2.1 Q-Learning
			11.2.2 R-Learning
		11.3 Model-Based RL
			11.3.1 SARSA Learning
			11.3.2 Dyna-Q Learning
			11.3.3 Temporal Difference
				11.3.3.1 TD(0) Algorithm
				11.3.3.2 TD(1) Algorithm
				11.3.3.3 TD(.) Algorithm
			11.3.4 Monte Carlo Method
				11.3.4.1 Monte Carlo Reinforcement Learning
				11.3.4.2 Monte Carlo Policy Evaluation
				11.3.4.3 Monte Carlo Policy Improvement
		11.4 Conclusion
		References
	Chapter 12 Application of Big Data and Machine Learning
		12.1 Introduction
		12.2 Motivation
		12.3 Related Work
		12.4 Application of Big Data and ML
			12.4.1 Healthcare
			12.4.2 Banking and Insurance
			12.4.3 Transportation
			12.4.4 Media and Entertainment
			12.4.5 Education
			12.4.6 Ecosystem Conservation
			12.4.7 Manufacturing
			12.4.8 Agriculture
		12.5 Issues and Challenges
		12.6 Conclusion
		References
Section 4: Machine Learning’s Next Frontier
	Chapter 13 Transfer Learning
		13.1 Introduction
			13.1.1 Motivation, Definition, and Representation
		13.2 Traditional Learning vs. Transfer Learning
		13.3 Key Takeaways: Functionality
		13.4 Transfer Learning Methodologies
		13.5 Inductive Transfer Learning
		13.6 Unsupervised Transfer Learning
		13.7 Transductive Transfer Learning
		13.8 Categories in Transfer Learning
		13.9 Instance Transfer
		13.10 Feature Representation Transfer
		13.11 Parameter Transfer
		13.12 Relational Knowledge Transfer
		13.13 Relationship With Deep Learning
			13.13.1 Transfer Learning in Deep Learning
			13.13.2 Types of Deep Transfer Learning
			13.13.3 Adaptation of Domain
			13.13.4 Domain Confusion
			13.13.5 Multitask Learning
			13.13.6 One-Shot Learning
			13.13.7 Zero-Shot Learning
		13.14 Applications: Allied Classical Problems
			13.14.1 Transfer Learning for Natural Language Processing
			13.14.2 Transfer learning for Computer Vision
			13.14.3 Transfer Learning for Audio and Speech
		13.15 Further Advancements and Conclusion
		References
Section 5: Hands-On and Case Study
	Chapter 14 Hands on MAHOUT—Machine Learning Tool
		14.1 Introduction to Mahout
			14.1.1 Features
			14.1.2 Advantages
			14.1.3 Disadvantages
			14.1.4 Application
		14.2 Installation Steps of Apache Mahout Using Cloudera
			14.2.1 Installation of VMware Workstation
			14.2.2 Installation of Cloudera
			14.2.3 Installation of Mahout
			14.2.4 Installation of Maven
			14.2.5 Testing Mahout
		14.3 Installation Steps of Apache Mahout Using Windows 10
			14.3.1 Installation of Java
			14.3.2 Installation of Hadoop
			14.3.3 Installation of Mahout
			14.3.4 Installation of Maven
			14.3.5 Path Setting
			14.3.6 Hadoop Configuration
		14.4 Installation Steps of Apache Mahout Using Eclipse
			14.4.1 Eclipse Installation
			14.4.2 Installation of Maven Through Eclipse
			14.4.3 Maven Setup for Mahout Configuration
			14.4.4 Building the Path-
			14.4.5 Modifying the pom.xml File
			14.4.6 Creating the Data File
			14.4.7 Adding External Jar Files
			14.4.8 Creating the New Package and Classes
			14.4.9 Result
		14.5 Mahout Algorithms
			14.5.1 Classification
			14.5.2 Clustering [36, 38, 39]
			14.5.3 Recommendation
		14.6 Conclusion
		References
	Chapter 15 Hands-On H2O Machine Learning Tool
		15.1 Introduction
		15.2 Installation
			15.2.1 The Process of Installation
		15.3 Interfaces
		15.4 Programming Fundamentals
			15.4.1 Data Manipulation
				15.4.1.1 Data Types
				15.4.1.2 Data Import
			15.4.2 Models
				15.4.2.1 Model Training
			15.4.3 Discovering Aspects
				15.4.3.1 Converting Data Frames
			15.4.4 H2O Cluster Actions
				15.4.4.1 H2O Key Value Retrieval
				15.4.4.2 H2O Cluster Connection
			15.4.5 Commands
				15.4.5.1 Cluster Information
				15.4.5.2 General Data Operations
				15.4.5.3 String Manipulation Commands
		15.5 Machine Learning in H2O
			15.5.1 Supervised Learning
			15.5.2 Unsupervised Learning
		15.6 Applications of H2O
			15.6.1 Deep Learning
			15.6.2 K-Fold Cross-Authentication or Validation
			15.6.3 Stacked Ensemble and Random Forest Estimator
		15.7 Conclusion
		References
	Chapter 16 Case Study: Intrusion Detection System Using Machine Learning
		16.1 Introduction
			16.1.1 Components Used to Design the Scenario Include
				16.1.1.1 Black Hole
				16.1.1.2 Intrusion Detection System
				16.1.1.3 Components Used From MATLAB Simulator
		16.2 System Design
			16.2.1 Three Sub-Network Architecture
			16.2.2 Using Classifiers of MATLAB
		16.3 Existing Proposals
		16.4 Approaches Used in Designing the Scenario
			16.4.1 Algorithm Used in QualNet
			16.4.2 Algorithm Applied in MATLAB
		16.5 Result Analysis
			16.5.1 Results From QualNet
				16.5.1.1 Deployment
				16.5.1.2 Detection
				16.5.1.3 Avoidance
				16.5.1.4 Validation of Conclusion
			16.5.2 Applying Results to MATLAB
				16.5.2.1 K-Nearest Neighbor
				16.5.2.2 SVM
				16.5.2.3 Decision Tree
				16.5.2.4 Naïve Bayes
				16.5.2.5 Neural Network
		16.6 Conclusion
		References
	Chapter 17 Inclusion of Security Features for Implications of Electronic Governance Activities
		17.1 Introduction
		17.2 Objective of E-Governance
		17.3 Role of Identity in E-Governance
			17.3.1 Identity
			17.3.2 Identity Management and its Buoyancy Against Identity Theft in E-Governance
		17.4 Status of E-Governance in Other Countries
			17.4.1 E-Governance Services in Other Countries Like Australia and South Africa
			17.4.2 Adaptation of Processes and Methodology for Developing Countries
			17.4.3 Different Programs Related to E-Governance
		17.5 Pros and Cons of E-Governance
		17.6 Challenges of E-Governance in Machine Learning
		17.7 Conclusion
		References
Index
EULA