Handbook of Machine Learning for Computational Optimization: Applications and Case Studies (Demystifying Technologies for Computational Excellence)

✍ Scribed by Vishal Jain (editor), Sapna Juneja (editor), Abhinav Juneja (editor), Ramani Kannan (editor)

• Publisher: CRC Press
• Year: 2021
• Tongue: English
• Leaves: 295
• Edition: 1
• Category: Library

✦ Synopsis

Technology is moving at an exponential pace in this era of computational intelligence. Machine learning has emerged as one of the most promising tools used to challenge and think beyond current limitations. This handbook will provide readers with a leading edge to improving their products and processes through optimal and smarter machine learning techniques.

This handbook focuses on new machine learning developments that can lead to newly developed applications. It uses a predictive and futuristic approach, which makes machine learning a promising tool for processes and sustainable solutions. It also promotes newer algorithms that are more efficient and reliable for new dimensions in discovering other applications, and then goes on to discuss the potential in making better use of machines in order to ensure optimal prediction, execution, and decision-making.

Individuals looking for machine learning-based knowledge will find interest in this handbook. The readership ranges from undergraduate students of engineering and allied courses to researchers, professionals, and application designers.

✦ Table of Contents

Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Preface
Editors
Contributors
Chapter 1 Random Variables in Machine Learning
1.1 Introduction
1.2 Random Variable
1.2.1 Definition and Classification
1.2.1.1 Applications in Machine Learning
1.2.2 Describing a Random Variable in Terms of Probabilities
1.2.2.1 Ambiguity with Reference to Continuous Random Variable
1.2.3 Probability Density Function
1.2.3.1 Properties of pdf
1.2.3.2 Applications in Machine Learning
1.3 Various Random Variables Used in Machine Learning
1.3.1 Continuous Random Variables
1.3.1.1 Uniform Random Variable
1.3.1.2 Gaussian (Normal) Random Variable
1.3.2 Discrete Random Variables
1.3.2.1 Bernoulli Random Variable
1.3.2.2 Binomial Random Variable
1.3.2.3 Poisson Random Variable
1.4 Moments of Random Variable
1.4.1 Moments about Origin
1.4.1.1 Applications in Machine Learning
1.4.2 Moments about Mean
1.4.2.1 Applications in Machine Learning
1.5 Standardized Random Variable
1.5.1 Applications in Machine Learning
1.6 Multiple Random Variables
1.6.1 Joint Random Variables
1.6.1.1 Joint Cumulative Distribution Function (Joint CDF)
1.6.1.2 Joint Probability Density Function (Joint pdf)
1.6.1.3 Statistically Independent Random Variables
1.6.1.4 Density of Sum of Independent Random Variables
1.6.1.5 Central Limit Theorem
1.6.1.6 Joint Moments of Random Variables
1.6.1.7 Conditional Probability and Conditional Density Function of Random Variables
1.7 Transformation of Random Variables
1.7.1 Applications in Machine Learning
1.8 Conclusion
References
Chapter 2 Analysis of EMG Signals using Extreme Learning Machine with Nature Inspired Feature Selection Techniques
2.1 Introduction
2.2 Data Set
2.3 Feature Extraction
2.4 Nature Inspired Feature Selection Methods
2.4.1 Particle Swarm Optimization Algorithm (PSO)
2.4.2 Genetic Algorithm (GA)
2.4.3 Fire-Fly Optimization Algorithm (FA)
2.4.4 Bat Algorithm (BA)
2.4.5 Whale Optimization Algorithm (WOA)
2.4.5.1 Exploitation Phase
2.4.5.2 Exploration Phase
2.5 Extreme Learning Machine (ELM)
2.6 Results and Discussion
2.7 Conclusion
References
Chapter 3 Detection of Breast Cancer by Using Various Machine Learning and Deep Learning Algorithms
3.1 Introduction
3.1.1 Risk Factors for Breast Cancer
3.1.2 Screening Guidelines
3.1.3 Consequences of Misidentifying the Tumor
3.1.4 Materials and Methods
3.2 Model Selection
3.2.1 Logistic Regression
3.2.2 Nearest Neighbor
3.2.3 Support Vector Machine
3.2.4 Naive Bayes Algorithm
3.2.5 Decision Tree Algorithm
3.2.6 Random Forest Classification
3.3 Detection of Breast Cancer by Using Deep Learning
3.4 Conclusion
References
Chapter 4 Assessing the Radial Efficiency Performance of Bus Transport Sector Using Data Envelopment Analysis
4.1 Introduction
4.1.1 Background Work
4.2 Methodology Framework
4.2.1 DEA Background
4.2.2 New Slack Model
4.3 Performance Evaluation of Depots
4.3.1 Data Collection
4.3.2 Region-wise Classification of Depots
4.3.3 Input and Output Parameters
4.3.4 Empirical Results
4.3.5 Input Targets for Inefficient Depots
4.4 Conclusion
Acknowledgement
References
Appendix (A)
Chapter 5 Weight-Based Codes—A Binary Error Control Coding Scheme—A Machine Learning Approach
5.1 Introduction
5.2 Encoding
5.3 Decoding (Machine Learning Approach)
5.3.1 Principle of Decoding
5.3.2 Algorithm
5.4 Output Test Case
5.5 Conclusion
References
Chapter 6 Massive Data Classification of Brain Tumors Using DNN: Opportunity in Medical Healthcare 4.0 through Sensors
6.1 Introduction
6.1.1 Brain Tumor
6.1.2 Big Data Analytics in Health Informatics
6.1.3 Machine Learning (ML) in Healthcare
6.1.4 Sensors for Internet of Things
6.1.5 Challenges and Critical Issues of IoT in Healthcare
6.1.6 Machine Learning (ML) and Artificial Intelligence (AI) for Health Informatics
6.1.7 Health Sensor Data Management
6.1.8 Multimodal Data Fusion for Healthcare
6.1.9 Heterogeneous Data Fusion and Context-Aware Systems—a Context-Aware Data Fusion Approach for Health-IoT
6.1.10 Role of Technology in Addressing the Problem of Integration of Healthcare System
6.2 Literature Survey
6.3 System Design and Methodology
6.3.1 System Design
6.3.2 CNN Architecture
6.3.3 Block Diagram
6.3.4 Algorithm(s)
6.3.5 Our Experimental Results, Interpretation, and Discussion
6.3.6 Implementation Details
6.3.7 Snapshots of Interfaces
6.3.8 Performance Evaluation
6.3.9 Comparison with Other Algorithms
6.4 Novelty in Our Work
6.5 Future Scope, Possible Applications, and Limitations
6.6 Recommendations and Consideration
6.7 Conclusions
References
Chapter 7 Deep Learning Approach for Traffic Sign Recognition on Embedded Systems
7.1 Introduction
7.2 Literature Review
7.3 General Challenges
7.4 Proposed Solution
7.4.1 Hardware
7.5 Models
7.5.1 YOLOV3
7.5.2 Tiny-YOLOV3
7.5.3 Darknet Reference Model
7.6 Flowcharts
7.7 Key Features of the System
7.8 Technology Stack
7.9 Dataset
7.9.1 Labeling/Annotating the Dataset
7.10 Training the Model
7.11 Result
7.12 Future Scope
References
Chapter 8 Lung Cancer Risk Stratification Using ML and AI on Sensor-Based IoT: An Increasing Technological Trend for Health of Humanity
8.1 Introduction
8.1.1 Motivation to the Study
8.1.2 Problem Statements
8.1.3 Authors’ Contributions
8.1.4 Research Manuscript Organization
8.1.5 Definitions
8.1.6 Computer-aided Diagnosis System (CADe or CADx)
8.1.7 Sensors for the Internet of Things
8.1.8 Wireless and Wearable Sensors for Health Informatics
8.1.9 Remote Human’s Health and Activity Monitoring
8.1.10 Decision-Making Systems for Sensor Data
8.1.11 Artificial Intelligence (AI) and Machine Learning (ML) for Health Informatics
8.1.12 Health Sensor Data Management
8.1.13 Multimodal Data Fusion for Healthcare
8.1.14 Heterogeneous Data Fusion and Context-Aware Systems—a Context-Aware Data Fusion Approach for Health-IoT
8.2 Literature Review
8.3 Proposed Systems
8.3.1 Framework or Architecture of the Work
8.3.2 Model Steps and Parameters
8.3.3 Discussions
8.4 Experimental Results and Analysis
8.4.1 Tissue Characterization and Risk Stratification
8.4.2 Samples of Cancer Data and Analysis
8.5 Novelties
8.6 Future Scope, Limitations, and Possible Applications
8.7 Recommendations and Considerations
8.8 Conclusions
References
Chapter 9 Statistical Feedback Evaluation System
9.1 Introduction
9.2 Related Work
9.3 Types of Feedback Evaluation Systems
9.3.1 Questionnaire-Based Feedback Evaluation System (QBFES)
9.3.2 Star-Point-based Feedback Evaluation System (SBFES)
9.3.3 Text-Based Feedback Evaluation System (TBFES)
9.4 Statistical Feedback Evaluation System
9.4.1 Aspect Extraction
9.4.1.1 Feedback Collector
9.4.1.2 Feedback Preprocessor
9.4.1.3 Aspect Validator
9.4.2 Aspect Weight Estimation
9.4.3 Sentiment Evaluation
9.4.3.1 Sentiment Estimator
9.4.3.2 Sentiment Aggregator
9.4.4 Customized Evaluation
9.4.5 Aspect-Based Questionnaire Design
9.5 Result Analysis and Discussion
9.6 Conclusion
9.7 Future Work
References
Chapter 10 Emission of Herbal Woods to Deal with Pollution and Diseases: Pandemic-Based Threats
10.1 Introduction
10.1.1 Scenario of Pollution and Need to Connect with Indian Culture
10.1.2 Global Pollution Scenario
10.1.3 Indian Crisis on Pollution and Worrying Stats
10.1.4 Efforts Made to Curb Pollution World Wide
10.1.5 Indian Ancient Vedic Sciences to Curb Pollution and Related Diseases
10.1.6 The Yajna Science: A Boon to Human Race from Rishis and Munis
10.1.7 The Science of Mantra Associated with Yajna and Its Scientific Effects
10.1.8 Effect of Different Woods and Cow Dung Used in Yajna
10.1.9 Use of Sensors and IoT to Record Experimental Data
10.1.10 Analysis and Pattern Recognition by ML and AI
10.2 Literature Survey
10.2.1 Gist
10.2.2 Methodology Used in This Paper
10.2.3 Instruments and Data Set Used
10.2.4 The Future Scope Discussed
10.3 The Methodology and Protocols Followed
10.4 Experimental Setup of an Experiment
10.4.1 Airveda and Different Sensor-Based Instruments
10.5 Results and Discussions
10.5.1 Mango v/s Banyan (Bargad)
10.5.1.1 Mango
10.5.1.2 Bargad
10.6 Applications of Yagya and Mantra Therapy in Pollution Control and its Significance
10.7 Future Research Perspectives
10.8 Novelty of Our Research
10.9 Recommendations
10.10 Conclusions
References
Chapter 11 Artificial Neural Networks: A Comprehensive Review
11.1 Introduction
11.2 Activation Function
11.2.1 Linear Activation Function
11.2.2 Nonlinear Activation Function
11.2.2.1 Sigmoid (Logistic) Function
11.2.2.2 Tanh Activation Function
11.2.2.3 Rectified Linear Unit (ReLU) Function
11.3 Artificial Neural Network (ANN)
11.3.1 Supervised Learning
11.3.2 Unsupervised Learning
11.3.3 Reinforcement Learning
11.4 Types of Artificial Neural Network
11.4.1 Single-Layer Feedforward Neural Network
11.4.2 Multilayer Feedforward Neural Networks
11.4.3 Recursive Neural Network (RNN)
11.4.4 Convolutional Layer Network (CNN)
11.4.5 Backpropagation Neural Network
11.4.5.1 Static Backpropagation
11.4.5.2 Recurrent Backpropagation
11.5 Problems in Artificial Neural Networks
11.5.1 Techniques to Avoid Overfitting When Neural Networks are Trained
11.6 Convergence of Neural Network
11.6.1 Adaptive Convergence (or Just Convergence)
11.6.2 Reactive Convergence
11.7 Key Features of the Error Surface
11.7.1 Local Minima
11.7.2 Flat Regions (Saddle Points)
11.7.3 High-Dimensional
11.8 Application of Artificial Neural Network
11.9 Conclusion
References
Chapter 12 A Case Study on Machine Learning to Predict the Students’ Result in Higher Education
12.1 Introduction
12.1.1 Literature Review
12.2 Proposed Model
12.2.1 Participants and Datasets
12.2.2 Data Retrieval
12.2.3 Data Preprocessing
12.3 Result and Discussion
12.3.1 Model Evaluation Metrics
12.3.2 Decision Tree Classification
12.3.3 KNN Classification
12.3.4 Random Forest Tree Classification
12.3.5 X-Gradient Boosting Tree Classification
12.4 Comparative Results for Different Classification Models
12.5 Conclusion and Future Scope
References
Chapter 13 Data Analytic Approach for Assessment Status of Awareness of Tuberculosis in Nigeria
13.1 Introduction
13.2 Related Works
13.3 Materials and Methods
13.3.1 Population and Sample
13.3.2 Tools and Designing
13.3.3 Task Procedures
13.3.4 Data Analysis and Results
13.4 Results and Discussion
13.5 Conclusions
Acknowledgements
References
Chapter 14 Active Learning from an Imbalanced Dataset: A Study Conducted on the Depression, Anxiety, and Stress Dataset
14.1 Introduction
14.2 Literature Survey
14.3 Problem Statement
14.4 Necessity of Defining the Problem/Research Gap
14.5 Objectives
14.5.1 Primary Objective
14.5.2 Secondary Objective
14.6 Dataset
14.6.1 Data Collection
14.6.2 Data Description
14.6.3 Data Preprocessing
14.6.4 Exploratory Data Analysis
14.6.4.1 Analysis of DASS
14.6.4.2 Analysis of the TIPI Test
14.6.4.3 Analysis of Time Taken by the Users to Complete the Survey
14.6.4.4 Analysis of the Validity-Check List and their Relationship with the Education Information
14.7 Implementation Design
14.7.1 Class Imbalance
14.7.2 SMOTE
14.7.3 Model Building
14.7.4 Evaluation Metric
14.8 Results and Conclusion
References
Chapter 15 Classification of the Magnetic Resonance Imaging of the Brain Tumor Using the Residual Neural Network Framework
15.1 Introduction
15.2 Literature Review
15.3 Architecture of Resnet Medical Imaging Modalities
15.4 Stages for Implementation of the Resnet Framework
15.4.1 Preprocessing
15.4.2 Training the Network
15.4.3 Segmentation
15.4.4 Focal Loss Function
15.5 Results and Discussions
15.6 Conclusions and Future Scope
References
Index