Machine Learning: Concepts, Techniques and Applications

Cover
Half Title
Title Page
Copyright Page
Table of Contents
Preface
Author Biography
Chapter 1: Introduction
1.1 Introduction
1.1.1 Intelligence
1.1.2 Learning
1.1.3 Informal Introduction to Machine Learning
1.1.4 Artificial Intelligence, Machine Learning, and Deep Learning
1.2 Need for Machine Learning
1.2.1 Extracting Relevant Information
1.2.2 Why Study Machine Learning?
1.2.3 Importance of Machine Learning
1.3 Machine Learning—Road Map
1.3.1 Early History
1.3.2 Focus on Neural Networks
1.3.3 Discovery of Foundational Ideas of Machine Learning
1.3.4 Machine Learning from Knowledge-Driven to Data-Driven
1.3.5 Applied Machine Learning—Text and Vision and Machine Learning Competitions
1.3.6 Deep Learning—Coming of Age of Neural Nets
1.3.7 Industrial Participation and Machine Learning
1.4 What Is Machine Learning?
1.5 Explaining Machine Learning
1.6 Areas of Influence for Machine Learning
1.7 Applications of Machine Learning
1.7.1 Applications of Machine Learning Across the Spectrum
1.7.2 Machine Learning in the Big Data Era
1.7.3 Interesting Applications
1.8 Identifying Problems Suitable for Machine Learning
1.9 Advantages of Machine Learning
1.10 Disadvantages of Machine Learning
1.11 Challenges of Machine Learning
1.12 Summary
1.13 Points to Ponder
E.1.3 Match the Columns
E.1.4 Sequencing
References
Chapter 2: Understanding Machine Learning
2.1 General Architecture of a Machine Learning System
2.2 Machine Learning Terminology
2.3 Types of Machine Learning Tasks
2.4 Design of a Simple Machine Learning System
2.4.1 Important Aspects in the Design of a Learning System
2.4.2 Illustrative Examples of the Process of Design
2.4.2.1 Recognition of Handwritten Characters
2.4.2.2 Checkers Learning
2.5 Summary
2.6 Points to Ponder
E.2 Exercises
E.2.1 Suggested Activities
Self-Assessment Questions
E.2.2 Multiple Choice Questions
E.2.3 Match the Columns
E.2.4 Short Questions
References
Chapter 3: Mathematical Foundations and Machine Learning
3.1 Introduction
3.2 What Is Linear Algebra?
3.2.1 Linear Algebra and Machine Learning
3.2.2 Matrix and Matrix Operations
3.2.2.1 Vector and Vector Operations
3.2.2.2 Operations with Vectors
3.2.2.3 Linear Dependence and Independence
3.2.2.3.1 Vector Projection
3.3 Probability Theory
3.3.1 Machine Learning and Probability Theory
3.3.2 Basics of Probability
3.3.3 Three Approaches to Probability
3.3.4 Types of Events in Probability
3.3.4.1 Visualizing Events in Sample Space
3.3.5 Probability of Multiple Random Variables
3.3.5.1 Simple Versus Joint Probability
3.3.6 Marginalization
3.3.7 Conditional Probability and Bayes Theorem
3.3.7.1 Bayes Theorem
3.3.7.2 Example 3.7 Bayes Theorem
3.3.7.3 Example 3.8 Bayes Theorem
3.3.8 Bayes Theorem and Machine Learning
3.3.9 Probability of Continuous Variables for Modelling the World
3.3.9.1 Characteristics of a Normal Probability Distribution
3.3.9.2 Standard Normal Probability Distribution
3.3.10 Use Case—Bayes Theorem—Diagnostic Test Scenario
3.4 Information Theory
3.4.1 Shannon’s Fundamental Theorems
3.4.1.1 Information Source
3.4.1.2 Stochastic Sources
3.4.2 Self Information
3.4.3 Entropy
3.4.4 Entropy for Memory and Markov Sources
3.4.4.1 The Source Coding Theorem
3.4.5 Cross Entropy
3.4.6 Kullback–Leibler Divergence or Relative Entropy
3.5 Summary
3.6 Points to Ponder
E.3 Exercises
E.3.1 Suggested Activities
Case study
Self-Assessment Questions
E.3.2 Multiple Choice Questions
E.3.3 Match the Columns
E.3.4 Problems
E.3.5 Short Questions
Chapter 4: Foundations and Categories of Machine Learning Techniques
4.1 Introduction
4.1.1 Data and Its Importance
4.1.2 Problem Dimensions
4.2 Data and Data Representation
4.2.1 Types of Data
4.2.2 Data Dependencies
4.2.3 Data Representation
4.2.4 Processing Data
4.2.5 Data Biases
4.2.6 Features and Feature Selection
4.2.6.1 Why Feature Selection?
4.2.6.2 Approaches to Feature Selection
4.2.6.3 Feature Extraction
4.3 Basis of Machine Learning
4.3.1 Inductive Learning
4.3.2 Generalization
4.3.3 Bias and Variance
4.3.4 Overfitting and Underfitting
4.3.4.1 Overfitting
4.3.4.1.1 Methods to Avoid Overfitting
4.3.4.2 Underfitting
4.4 Issues in Building Machine Learning Models
4.5 Offline and Online Machine Learning
4.6 Underlying Concepts of Machine Learning Algorithms—Parametric and Nonparametric Algorithms
4.6.1 Parametric Learning Versus Nonparametric
4.7 Approaches to Machine Learning Algorithms
4.8 Types of Machine Learning Algorithms
4.8.1 Supervised Learning
4.8.1.1 Workflow of a Supervised Learning System
4.8.1.2 Classification and Regression
4.8.1.3 Examples of Supervised Learning
4.8.2 Unsupervised Learning
4.8.2.1 Workflow of Unsupervised Learning System
4.8.2.2 Clustering, Association Rule Mining, and Dimensionality Reduction
4.8.3 Semi-supervised Learning
4.8.4 Reinforcement Learning
4.8.5 Neural Networks and Deep Learning
4.9 Summary
4.10 Points to Ponder
E.4 Exercises
E.4.1 Suggested Activities
Use Case
Thinking Exercise
Self-Assessment Questions
E.4.2 Multiple Choice Questions
E.4.3 Match the Columns
E.4.4 Short Questions
Chapter 5: Machine Learning: Tools and Software
5.1 Weka Tool
5.1.1 Features of Weka
5.1.2 Installation of Weka
5.1.3 Weka Application Interfaces
5.1.4 Data Formats of Weka
5.1.5 Weka Explorer
5.1.6 Data Preprocessing
5.1.7 Understanding Data
5.1.7.1 Selecting Attributes
5.1.7.2 Removing Attributes
5.1.7.3 Applying Filters
5.2 Introduction to Machine Learning Support in R
5.2.1 Advantages to Implement Machine Learning Using R Language
5.2.2 Popular R Language Packages
5.2.3 Application of R in Machine Learning
5.2.4 Examples of Machine Learning Problems
5.2.5 Types of Machine Learning Problems
5.2.6 Setting Up Environment for Machine Learning with R Programming Using Anaconda
5.2.7 Running R Commands
5.2.8 Installing Machine Learning Packages in R
5.2.9 Specific Machine Learning Packages in R
5.2.10 Supervised Learning Using R
5.2.11 Unsupervised Learning Using R
5.2.11.1 Implementing k - Means Clustering in R
5.3 Summary
5.4 Points to Ponder
E.5 Exercises
E.5.1. Suggested Activities
Self-Assessment Questions
E.5.2 Multiple Choice Questions
E.5.3 Match the Columns
E.5.4 Short Questions
Chapter 6: Classification Algorithms
6.1 Introduction
6.1.1 Basic Concepts of Classification
6.1.2 Binary Classification
6.1.3 Multi-Class Classification
6.1.4 Multi-Label Classification for Machine Learning
6.2 Decision Based Methods—Nonlinear Instance-Based Methods— k- Nearest Neighbor
6.2.1 Introduction
6.2.2 Need for KNN
6.2.3 Working of KNN
6.2.4 Calculating the Distance
6.2.5 Two Parameters of KNN
6.2.6 KNN Algorithm
6.2.7 Pros and Cons of KNN
6.3 Decision Based Methods—Decision Tree Algorithm
6.3.1 Decision Tree Based Algorithm
6.3.2 Terminology Associated with Decision Trees
6.3.3 Assumptions While Creating Decision Tree
6.3.4 How Do Decision Trees Work?
6.3.5 Steps in ID3 Algorithm
6.3.6 Attribute Selection Measures
6.3.6.1 Entropy
6.3.6.2 Gini Impurity
6.3.6.3 Information Gain
6.3.6.3.1 Information Gain—An Example
6.3.6.4 Calculation of the Entropies
6.3.6.5 Computing the Gain
6.3.6.6 Information Gain Versus Gini Impurity
6.4 Linear Models—Support Vector Machines
6.4.1 Nonlinear Data
6.4.2 Working of SVM
6.4.3 Important Concepts in SVM
6.4.3.1 Support Vectors
6.4.3.2 Hard Margin
6.4.3.3 Soft Margin
6.4.3.4 Different Kernels
6.4.4 Tuning Parameters
6.4.5 Advantages and Disadvantages of SVM
6.5 Use Cases
6.5.1 Healthcare Industries
6.5.2 Banking Sectors
6.5.3 Educational Sectors
6.6 Summary
6.7 Points to Ponder
E.6 Exercises
E.6.1 Suggested Activities
E.6.2 Self-Assessment Questions
E.6.3 Multiple Choice Questions
E.6.4 Match the Columns
E.6.5 Short Questions
Chapter 7: Probabilistic and Regression Based Approaches
7.1 Probabilistic Methods
7.1.1 Introduction—Bayes Learning
7.1.2 Bayesian Learning
7.1.3 Interpretation of Bayes Rule
7.1.4 Benefits of Bayesian Learning
7.1.5 Problems with Bayesian Learning
7.2 Algorithms Based on Bayes Framework
7.2.1 Choosing Hypotheses
7.2.2 Bayesian Classification
7.2.2.1 Discriminative Model
7.2.2.2 Generative Model
7.3 Naïve Bayes Classifier
7.3.1 Naïve Bayes Classifier: Assumptions
7.3.2 Naïve Bayes Algorithm
7.3.3 Characteristics of Naïve Bayes
7.3.4 Naïve Bayes Classification—Example 1—Corona Dataset
7.4 Bayesian Networks
7.4.1 Foundations of Bayesian Network
7.4.2 Bayesian Network Perspectives
7.4.3 Bayesian Network—Probability Fundamentals
7.4.4 Semantics of Bayesian Networks
7.4.5 Bayesian Network—Putting It All Together
7.4.5.1 Independence
7.4.5.2 Putting It Together
7.4.6 Limitations of Bayesian Networks
7.4.7 Constructing Bayesian Networks
7.4.8 Bayesian Networks—Eating Habits
7.4.9 Causes and Bayes’ Rule
7.4.10 Conditional Independence in BNs
7.4.11 Bayesian Networks—Alarm (from Judea Pearl)— Example 2
7.4.11.1 Semantics of Bayesian Networks—Alarm Network
7.4.11.2 Inferences in Bayesian Networks—Alarm Network
7.5 Regression Methods
7.5.1 Linear Regression Models
7.5.1.1 Steps in Learning a Linear Regression Model
7.5.1.1.1 Choosing the Model
7.5.1.1.2 Defining Loss Function
7.5.1.1.3 Controlling Model Complexity and Overfitting
7.5.1.1.4 Fitting or Optimizing the Model—Gradient Descent
7.5.2 Logistic Regression
7.6 Summary
7.7 Points to Ponder
E.7 Exercise
E.7.1 Suggested Activities
Self-Assessment Questions
E.7.2 Multiple Choice Questions
E.7.3 Match the Columns
E.7.4 Problems
E.7.5 Short Questions
References
Chapter 8: Performance Evaluation and Ensemble Methods
8.1 Introduction
8.2 Classification Metrics
8.2.1 Binary Classification
8.2.1.1 Accuracy
8.2.1.2 Sensitivity or True Positive Rate
8.2.1.3 Precision
8.2.1.4 Precision/Recall Trade-off
8.2.1.5 F1 Score
8.2.1.6 ROC/AUC Curve
8.2.2 Multi-Class Classification
8.3 Cross-Validation in Machine Learning
8.4 Ensemble Methods
8.4.1 Types of Ensemble Methods
8.4.1.1 Bagging
8.4.1.2 Comparison of Bagging and Boosting
8.4.1.3 Stacking
8.5 Handling Missing and Imbalanced Data
8.6 Summary
8.7 Points to Ponder
E.8 Exercises
E.8.1 Suggested Activities
Self-Assessment Questions
E.8.2 Multiple Choice Questions
E.8.3 Match the Columns
E.8.4 Short Questions
References
Chapter 9: Unsupervised Learning
9.1 Introduction to Unsupervised Learning
9.1.1 Importance of Unsupervised Learning
9.1.2 Challenges of Unsupervised Learning
9.2 Applications of Unsupervised Learning
9.3 Supervised Learning Versus Unsupervised Learning
9.4 Unsupervised Learning Approaches
9.5 Clustering
9.5.1 Clusters—Distance Viewpoint
9.5.2 Applications of Clustering
9.6 Similarity Measures
9.6.1 Distance Functions for Numeric Attributes
9.6.2 Distance Functions for Binary Attributes
9.7 Methods of Clustering
9.7.1 Hierarchical Clustering
9.7.2 Types of Hierarchical Clustering
9.7.2.1 Agglomerative Clustering
9.7.2.2 Divisive Clustering
9.7.3 Hierarchical Clustering: The Algorithm
9.7.3.1 Hierarchical Clustering: Merging Clusters
9.8 Agglomerative Algorithm
9.9 Issues Associated with Hierarchical Clustering
9.10 Partitional Algorithm
9.11 k - Means Clustering
9.11.1 Steps of k - Means
9.11.2 Issues Associated with the k - Means Algorithm
9.11.3 Evaluating k - Means Clusters
9.11.4 Strengths and Weaknesses of k - Means
9.12 Cluster Validity
9.12.1 Comparing with Ground Truth
9.12.2 Purity Based Measures
9.12.3 Internal Measure
9.13 Curse of Dimensionality
9.14 Dimensionality Reduction
9.15 The Process of Dimensionality Reduction
9.15.1 Criterion for Reduction
9.15.2 Feature Reduction and Feature Selection
9.16 Dimensionality Reduction with Feature Reduction
9.16.1 Principle Component Analysis (PCA)
9.16.1.1 PCA Methodology
9.16.2 Fisher Linear Discriminant Approach
9.16.2.1 Fisher Linear Discriminant
9.16.3 Singular Value Decomposition
9.17 Association Rule Mining
9.17.1 Market-Basket Analysis
9.17.2 Association Rule Mining—Basic Concepts
9.17.3 Apriori Algorithm
9.17.3.1 Apriori Principle
9.17.3.2 The Algorithm
9.17.3.3 Example of Frequent Itemset Generation Using Apriori Algorithm
9.17.3.4 Improving Apriori’s Efficiency
9.18 Summary
9.19 Points to Ponder
E.9 Exercises
E.9.1 Suggested Activities
Self-Assessment Questions
E.9.2 Multiple Choice Questions
E.9.3 Match the Columns
E.9.4 Problems
E.9.5 Short Questions
References
Chapter 10: Sequence Models
10.1 Sequence Models
10.2 Applications of Sequence Models
10.2.1 Examples of Applications with Sequence Data
10.2.2 Examples of Application Scenario of Sequence Models
10.3 Modelling Sequence Learning Problems
10.4 Markov Models
10.4.1 Types of Markov Models
10.4.2 Markov Chain Model
10.4.3 Hidden Markov Model (HMM)
10.4.3.1 Parameters of an HMM
10.4.3.2 The Three Problems of HMM
10.4.4 Markov Decision Process
10.4.5 Partially Observable Markov Decision Process
10.4.6 Markov Random Field
10.5 Data Stream Mining
10.5.1 Challenges in Data Stream Mining
10.6 Learning from Stream Data
10.6.1 Tree Based Learning
10.6.2 Adaptive Learning Using Naïve Bayes
10.6.3 Window Management Models
10.6.4 Data Stream Clustering
10.7 Applications
10.7.1 Applications of Markov Model
10.7.2 Applications of Stream Data Processing
10.7.2.1 Fraud and Anomaly Detection
10.7.2.2 Internet of Things (IoT) Edge Analytics
10.7.2.3 Customization for Marketing and Advertising
10.8 Summary
10.9 Points to Ponder
E.10 Exercises
E.10.1 Suggested Activities
Self-Assessment Questions
E.10.2 Multiple Choice Questions
E.10.3 Match the Columns
E.10.4 Problems
E.10.5 Short Questions
Chapter 11: Reinforcement Learning
11.1 Introduction
11.2 Action Selection Policies
11.3 Finite Markov Decision Processes
11.4 Problem Solving Methods
11.4.1 Dynamic Programming
11.4.2 Monte Carlo Methods
11.4.2.1 Monte Carlo Reinforcement Learning
11.4.2.2 Finding the Optimal Policy Using Monte Carlo
11.4.2.3 Monte Carlo with Fresh Exploration
11.4.2.4 Monte Carlo with Continual Exploration
11.4.3 Temporal Difference Learning
11.4.3.1 Q-learning
11.4.3.2 State–Action–Reward–State–Action (SARSA)
11.4.3.3 Deep Q-Networks (DQN)
11.5 Asynchronous Reinforcement Learning
11.5.1 Asynchronous Reinforcement Learning Procedure
11.6 Summary
11.7 Points to Ponder
E.11 Exercises
E.11.1 Suggested Activities
Self-Assessment Questions
E.11.2 Multiple Choice Questions
E.11.3 Match the Columns
E.11.4 Short Questions
Chapter 12: Machine Learning Applications: Approaches
12.1 Introduction
12.2 Machine Learning Life Cycle
12.3 Choosing a Machine Learning Algorithm for a Problem
12.4 Machine Learning and Its Applications
12.5 Machine Learning for Natural Language Processing
12.6 Recommendation Systems
12.6.1 Basics of Recommendation Systems
12.6.2 Utility Matrix and Recommendation Systems
12.6.3 Issues Associated with Recommendation Systems
12.6.3.1 Construction of Utility Matrix
12.6.3.2 The Cold Start Problem
12.6.3.3 Data Sparsity
12.6.3.4 Inability to Capture Changing User Behaviour
12.6.3.5 Other Issues
12.6.4 Types of Recommendation Algorithms
12.6.5 Collaborative Filtering: Memory Based Methods
12.6.5.1 Collaborative Filtering—Neighbor Based Algorithm
12.6.5.2 User Based Collaborative Filtering
12.6.5.3 Item Based Collaborative Filtering
12.6.5.4 Advantages and Disadvantages of Collaborative Based Filtering
12.6.6 Content Based Recommendation
12.6.6.1 Advantages and Disadvantages of Content Based Recommendation
12.7 Context Aware Recommendations
12.8 Summary
12.9 Points to Ponder
E.12 Exercises
E.12.1 Suggested Activities
Use case
Thinking Exercise
Self-Assessment Questions
E.12.2 Multiple Choice Questions
E.12.3 Match the Columns
E.12.4 Short Questions
Chapter 13: Domain Based Machine Learning Applications
13.1 Introduction
13.2 Everyday Examples of Machine Learning Applications
13.2.1 Personal Smart Assistants
13.2.2 Product Recommendation
13.2.3 Email Intelligence
13.2.4 Social Networking
13.2.5 Commuting Services
13.3 Machine Learning in Health Care
13.3.1 Why Machine Learning for Health Care Now?
13.3.2 What Makes Health Care Different?
13.3.3 Disease Identification and Diagnosis
13.3.4 Medical Imaging and Diagnosis
13.3.5 Smart Health Records
13.3.6 Drug Discovery and Manufacturing
13.3.7 Personalized Medicine and Treatment
13.3.8 Disease Outbreak Prediction
13.4 Machine Learning for Education
13.4.1 Personalized and Adaptive Learning
13.4.2 Increasing Efficiency
13.4.3 Learning Analytics
13.4.4 Predicative Analytics
13.4.5 Evaluating Assessments
13.5 Machine Learning in Business
13.5.1 Customer Service
13.5.2 Financial Management
13.5.3 Marketing
13.5.4 Consumer Convenience
13.5.5 Human Resource Management
13.6 Machine Learning in Engineering Applications
13.6.1 Manufacturing
13.6.2 Water and Energy Management
13.6.3 Environment Engineering
13.7 Smart City Applications
13.8 Summary
13.9 Points to Ponder
E.13 Exercises
E.13.1 Suggested Activities
Self-Assessment Questions
E.13.2 Multiple Choice Questions
E.13.3 Questions
Chapter 14: Ethical Aspects of Machine Learning
14.1 Introduction
14.2 Machine Learning as a Prediction Model
14.3 Ethics and Ethical Issues
14.3.1 Examples of Ethical Concerns
14.4 Ethics and Machine Learning
14.4.1 Opinions of Ethics in AI and Machine Learning
14.5 Fundamental Concepts of Responsible and Ethical Machine Learning
14.5.1 Current Legislation—General Data Protection Regulation (GDPR)
14.6 Fairness and Machine Learning
14.7 Bias and Discrimination
14.7.1 Examples of Bias and Dealing with Bias
14.7.1.1 Bias in Face Recognition Systems
14.7.1.2 Gender-Biased Issues in Natural Language Processing
14.7.1.3 Credit Score Computation
14.7.1.4 User Profiling and Personalization
14.7.2 Types of Bias
14.7.3 Data Bias and the Data Analysis Pipeline
14.8 Fairness Testing
14.8.1 Fairness and Evaluation Metrics
14.9 Case Study: LinkedIn Representative Talent Search
14.10 Explainability
14.10.1 Handling Black Box Machine Learning
14.10.2 Achieving Explainable Machine Learning
14.10.2.1 Attribution Methods
14.10.2.2 Other Methods for Explainability
14.10.3 Example: LinkedIn’s Approach to Explainability
14.11 Transparency
14.11.1 Issues of Transparency and Their Mitigation
14.12 Privacy
14.12.1 Data Privacy
14.12.2 Privacy Attacks
14.12.3 Privacy-Preserving Techniques
14.13 Summary
14.14 Points to Ponder
E.14 Exercises
E.14.1 Suggested Activities
Self-Assessment Questions
E.14.2 Multiple Choice Questions
E.14.3 Match the Columns
E.14.4 Short Questions
References
Chapter 15: Introduction to Deep Learning and Convolutional Neural Networks
15.1 Introduction
15.2 Example of Deep Learning at Work
15.3 Evolution of Deep Learning
15.4 Deep Learning in Action
15.4.1 Applications
15.4.2 Differences Between Machine Learning and Deep Learning
15.5 Neural Networks
15.5.1 Basic Components of Biological Neurons
15.5.1.1 Perceptrons
15.5.1.2 Perceptron Training
15.5.2 Activation Functions
15.6 Learning Algorithm
15.6.1 Backpropagation
15.6.2 Chain Rule
15.6.3 Backpropagation—For Outermost Layer
15.6.4 Backpropagation—For Hidden Layer
15.7 Multilayered Perceptron
15.7.1 The Basic Structure
15.7.2 Backpropagation
15.8 Convolutional Neural Network
15.8.1 Biological Connection to CNN
15.8.2 The Architecture of Convolutional Neural Networks (CNNs)
15.8.3 Convolutional Networks in the History of Deep Learning
15.8.4 Learning in Convolutional Neural Networks
15.8.4.1 Image Channels
15.8.4.2 Convolution
15.8.4.3 Pooling
15.8.4.4 Flattening
15.8.4.5 Full Connection: A Simple Convolutional Network
15.9 Summary
15.10 Points to Ponder
E.15 Exercises
E.15.1 Suggested Activities
Self-Assessment Questions
E.15.2 Multiple Choice Questions
E.15.3 Match the Columns
E.15.4 Short Questions
Chapter 16: Other Models of Deep Learning and Applications of Deep Learning
16.1 Recurrent Neural Networks (RNNs)
16.1.1 Working of RNN
16.1.2 Training Through RNN
16.2 Auto-encoders
16.2.1 Different Types of Auto-encoders
16.3 Long Short Term Memory Networks
16.3.1 How Does LSTM Solve the Problem of Vanishing and Exploring Gradients?
16.4 Generative Adversarial Networks (GANs)
16.4.1 Idea Behind GAN
16.4.2 Generative Versus Discriminative Models
16.5 Radial Basis Function Neural Network
16.5.1 Components of RBNN
16.6 Multilayer Perceptrons (MLPs)
16.6.1 Layers
16.6.2 Learning in MLP
16.6.3 Applications of MLPs
16.6.3.1 RBNN and MLP
16.6.3.2 Comparing RBNN and MLP
16.7 Self-Organizing Maps (SOMs)
16.7.1 Architecture SOM
16.7.2 Working of SOM
16.7.2.1 An Illustration of the Training of a Self-Organizing Map
16.7.3 Options for Initialization
16.7.3.1 Drawbacks to Kohonen Maps
16.8 Restricted Boltzmann Machines (RBMs)
16.8.1 Working of Restricted Boltzmann Machine
16.8.2 Advantages and Drawbacks of RBM
16.9 Deep Belief Networks (DBNs)
16.9.1 Creation of a Deep Belief Network
16.9.2 Benefits and Drawbacks of Deep Belief Networks
16.9.2.1 Benefits
16.9.2.2 Drawbacks
16.9.2.3 Applications of DBN
16.10 Applications
16.10.1 Deep Learning for Vision
16.10.2 Deep Learning for Drug Discovery
16.10.3 Deep Learning for Business and Social Network Analytics
16.11 Summary
16.12 Points to Ponder
E.16 Exercises
E.16.1 Suggested Activities
Self-Assessment Questions
E.16.2 Multiple Choice Questions
E.16.3 Match the Columns
E.16.4 Short Questions
References
A1: Solutions
Chapter – 1
Chapter – 2
Chapter – 3
Chapter – 4
Chapter – 5
Chapter – 6
Chapter – 7
Chapter – 8
Chapter – 9
Chapter – 10
Chapter – 11
Chapter – 12
Chapter – 13
Chapter – 14
Chapter – 15
Chapter – 16
Index