Design of Intelligent Applications using Machine Learning and Deep Learning Techniques
- Length: 446 pages
- Edition: 1
- Language: English
- Publisher: Chapman and Hall/CRC
- Publication Date: 2021-08-16
- ISBN-10: 0367679795
- ISBN-13: 9780367679798
- Sales Rank: #0 (See Top 100 Books)
Machine learning (ML) and deep learning (DL) algorithms are invaluable resources for Industry 4.0 and allied areas and are considered as the future of computing. A subfield called neural networks, to recognize and understand patterns in data, helps a machine carry out tasks in a manner similar to humans. The intelligent models developed using ML and DL are effectively designed and are fully investigated – bringing in practical applications in many fields such as health care, agriculture and security. These algorithms can only be successfully applied in the context of data computing and analysis. Today, ML and DL have created conditions for potential developments in detection and prediction.
Apart from these domains, ML and DL are found useful in analysing the social behaviour of humans. With the advancements in the amount and type of data available for use, it became necessary to build a means to process the data and that is where deep neural networks prove their importance. These networks are capable of handling a large amount of data in such fields as finance and images. This book also exploits key applications in Industry 4.0 including:
- Fundamental models, issues and challenges in ML and DL.
- Comprehensive analyses and probabilistic approaches for ML and DL.
- Various applications in healthcare predictions such as mental health, cancer, thyroid disease, lifestyle disease and cardiac arrhythmia.
- Industry 4.0 applications such as facial recognition, feather classification, water stress prediction, deforestation control, tourism and social networking.
- Security aspects of Industry 4.0 applications suggest remedial actions against possible attacks and prediction of associated risks.
- Information is presented in an accessible way for students, researchers and scientists, business innovators and entrepreneurs, sustainable assessment and management professionals.
This book equips readers with a knowledge of data analytics, ML and DL techniques for applications defined under the umbrella of Industry 4.0. This book offers comprehensive coverage, promising ideas and outstanding research contributions, supporting further development of ML and DL approaches by applying intelligence in various applications.
Cover Half Title Title Page Copyright Page Table of Contents Preface Editors Contributors 1. Data Acquisition and Preparation for Artificial Intelligence and Machine Learning Applications 1.1 Introduction 1.2 Reference Architecture 1.2.1 Data Sources 1.2.2 Data Storage 1.2.3 Batch Processing 1.2.4 Real-Time Message Ingestion 1.2.5 Stream Processing 1.2.6 Machine Learning 1.2.7 Analytical Data Store 1.2.8 Analytics and Reports 1.2.9 Orchestration 1.3 Data Acquisition Layer 1.3.1 File Systems 1.3.2 Databases 1.3.3 Applications 1.3.4 Devices 1.3.5 Enterprise Data Gateway 1.3.6 Field Gateway 1.3.7 Data Integration Services 1.3.8 Data Ingestion Services 1.4 Data Ingestion Layer 1.4.1 Data Storage Layer 1.4.2 Landing Layer 1.4.3 Cleansed Layer 1.4.4 Processed Layer 1.4.5 Data Processing Layer 1.4.6 Data Processing Engine 1.4.7 Data Processing Programs 1.4.8 Scheduling Engine 1.4.9 Scheduling Scripts 1.5 Data Quality and Cleansing Layer 1.5.1 Master Data Management (MDM)System 1.5.2 Master Data Management (MDM)Referencing Programs 1.5.3 Data Quality Check Programs 1.5.4 Rejected/Quarantined Layer Bibliography 2. Fundamental Models in Machine Learning and Deep Learning 2.1 Introduction 2.2 Classification of Machine Learning Models 2.2.1 Supervised Learning 2.2.2 Unsupervised Learning 2.2.3 Semi-Supervised Learning 2.2.4 Reinforcement Learning 2.3 Fundamental Supervised Learning Models 2.3.1 Regression 2.3.2 Classification 2.3.2.1 Logistic Regression 2.3.2.2 Support Vector Machines 2.3.3 Classification–Regression 2.3.3.1 Decision Tree 2.3.3.2 Random Forest 2.3.3.3 Artificial Neural Network 2.3.4 Implementation Code Snippet for Classification and Classification–Regression Techniques 2.4 Fundamental Unsupervised Learning Models 2.4.1 k-means Clustering 2.4.2 Apriori Algorithm 2.5 Fundamental Deep Learning Models 2.5.1 Autoencoder 2.5.2 Recurrent Neural Network 2.5.3 Convolutional Neural Network References 3. Research Aspects of Machine Learning: Issues, Challenges, and Future Scope 3.1 Introduction 3.1.1 The ML Approach 3.2 Issues 3.3 Challenges 3.3.1 ML Challenges Originating from BD 3.3.1.1 Volume 3.3.1.2 Variety 3.3.1.3 Velocity 3.3.1.4 Veracity 3.3.2 ML Challenges Originating from Wireless Sensor Networking 3.3.2.1 ML-Based 3.3.2.2 Infrastructure Update 3.3.2.3 ML-Based Network Slicing 3.3.2.4 Standard Datasets and Environments for Research 3.3.2.5 Theoretical Guidance for Algorithm Implementation 3.3.2.6 Transfer Learning 3.3.3 ML Challenges Originating from Blockchain 3.3.3.1 Suitability 3.3.3.2 Infrastructure 3.3.3.3 Privacy 3.3.3.4 Memory 3.3.3.5 Implementation 3.3.3.6 Security 3.3.3.7 Quantum Resilience 3.3.4 ML Challenges Originating from IoT 3.3.5 ML Challenges Originating from Bioinformatics 3.3.5.1 ML Algorithms for Bioinformatics 3.3.6 Accuracy 3.3.7 Complexity 3.3.7.1 Algorithm Computational Efficiency 3.4 Future Scope 3.4.1 Agriculture 3.4.2 Integrative Framework for Anticancer Drug Prediction 3.4.3 Medical Disease Diagnosis 3.4.4 Stock Market Analysis 3.4.5 IoT 3.4.6 Health Data Linkage 3.4.7 Deep Learning 3.4.8 Educational Technology References 4. Comprehensive Analysis of Dimensionality Reduction Techniques for Machine Learning Applications 4.1 Introduction 4.2 Missing Value Ratio 4.3 Low Variance Filter 4.4 High Correlation Filter 4.5 Principal Component Analysis 4.6 Independent Component Analysis 4.7 Backward Feature Elimination 4.8 Forward Feature Construction 4.9 Singular Value Decomposition 4.10 Random Forest 4.11 Conclusion References 5. Application of Deep Learning in Counting WBCs, RBCs, and Blood Platelets Using Faster Region-Based Convolutional Neural Network 5.1 Introduction 5.2 Convolutional Neural Network (CNN) 5.2.1 Activation Function 5.2.2 Convolutional Function 5.2.3 Pooling Function 5.2.4 Fully Connected Layer 5.3 Region-Based Convolutional Neural Network (RCNN) 5.3.1 RCNN Architecture 5.4 Fast RCNN 5.5 Faster RCNN 5.5.1 Region Proposal Network (RPN) 5.5.2 Faster RCNN Architecture 5.6 Implementation 5.6.1 Steps for Implementation 5.6.2 Some Problems, Solutions, and Suggestions 5.7 Viability of the Solution References 6. Application of Neural Network and Machine Learning in Mental Health Diagnosis 6.1 Introduction 6.2 Data Collection Approaches 6.2.1 Standard Questionnaire–Based Evaluation 6.2.1.1 Hamilton Depression Rating Scale 6.2.1.2 Beck’s Depression Inventory 6.2.1.3 Patient Health Questionnaire 6.2.2 Emotion Analysis from Speech Signals Using Convolutional Neural Networks 6.2.2.1 Working of Convolutional Neural Network Model 6.2.2.2 Result and Conclusion of the Experiment 6.2.3 Text-Based Emotion Recognition System 6.2.3.1 Formal Definition of TBERS 6.2.3.2 Different Approaches to Implement TBERS 6.3 Proposed Model 6.3.1 Model Architecture 6.3.2 Classifiers Used for Diagnosis 6.3.2.1 Logistic Model Tree Decision Tree Algorithm 6.3.2.2 Classification of Stress Recognition Using Artificial Neural Network 6.4 Therapy Using Intelligent System Games 6.5 Conclusion and Future Scope References 7. Application of Machine Learning in Cardiac Arrhythmia 7.1 Introduction 7.2 Proposed Machine Learning Implementation 7.2.1 Data Preparation 7.2.2 Data Preprocessing 7.2.3 Development of Model for Detection 7.2.4 Training and Experimenting the Model for the Data 7.2.5 Detection of Arrhythmia 7.3 Experimentation and Results 7.4 Evaluation Parameters and Measures 7.5 Comparison with Existing Systems 7.6 Conclusion References 8. Advances in Machine Learning and Deep Learning Approaches for Mammographic Breast Density Measurement for Breast Cancer Risk Prediction: An Overview 8.1 Introduction 8.2 Machine Learning Approach 8.2.1 Preprocessing 8.2.2 Breast Border Detection Algorithms 8.2.2.1 Gray-Level Thresholding 8.2.2.2 Iterative Optimal Thresholding 8.2.2.3 Otsu’s Optimal Thresholding 8.2.2.4 Minimum Cross-Entropy Thresholding 8.2.3 Pectoral Muscle Removal Algorithms 8.2.3.1 Image-Based Approach 8.2.3.2 Model-Based Approach 8.2.4 Segmentation 8.2.4.1 Region-Growing Algorithm 8.2.4.2 Graph-Cut Algorithm 8.2.4.3 Fuzzy C-Means 8.2.4.4 Watershed Algorithm 8.2.4.5 Otsu’s Optimal Thresholding 8.2.4.6 Fusion of K-Means and Region-Growing Algorithms 8.2.5 Feature Extraction and Classification 8.2.5.1 Statistical Feature Extraction 8.2.5.2 Feature Reduction 8.2.5.3 Classification 8.3 Deep Learning Approach 8.3.1 Preprocessing 8.3.2 Design of Convolutional Neural Network Algorithm 8.3.2.1 Input Layer 8.3.2.2 Convolutional Layer 8.3.2.3 Max Pooling Layers 8.3.2.4 Activation Function 8.3.2.5 Classification Layer 8.3.2.6 Dropout Layer 8.3.2.7 Kernel Selection 8.3.2.8 Development of Baseline (Density Map) for Breast Density Prediction 8.3.3 Validation and Testing of Machine and Deep Learning Model 8.3.4 Related Work 8.4 Findings and Discussion 8.5 Conclusion References 9. Applications of Machine Learning in Psychology and the Lifestyle Disease Diabetes Mellitus 9.1 Introduction 9.1.1 Application of Machine Learning in Psychology 9.1.1.1 Detecting Depression Using Machine Learning 9.1.2 Application of Machine Learning in Detecting Lifestyle Disease 9.1.2.1 Detecting Diabetic Retinopathy Using Machine Learning 9.2 Machine Learning for Depression Detection 9.2.1 Preprocessing 9.2.2 Algorithms Used 9.2.3 Score Generation 9.3 Machine Learning for Diabetic Retinopathy Detection 9.3.1 Preprocessing 9.3.2 Algorithms Used 9.3.2.1 Training Phase 9.3.2.2 Testing Phase 9.4 Conclusion References 10. Application of Machine Learning and Deep Learning in Thyroid Disease Prediction 10.1 Chapter Flow 10.2 Introduction 10.3 Related Work with Thyroid Prediction 10.4 Machine Learning Model for Thyroid Prediction 10.4.1 Supervised Learning 10.4.2 Unsupervised Learning 10.5 Implementation of Model for Thyroid Prediction 10.6 Impact/Case Study of Work 10.7 Advantages 10.8 Conclusion 10.9 Future Scope References 11. Application of Machine Learning in Fake News Detection 11.1 Introduction 11.1.1 Fake News: What It Is? 11.1.2 Essential Concept 11.1.2.1 Writing Mode–Based Fake News Analysis 11.1.2.2 Propagation-Based Fake News Analysis 11.1.2.3 User-Based Fake News Analysis 11.1.3 Types of Fake News 11.1.4 Overview of Chapter 11.2 Traditional Approaches for Fake News Detection 11.2.1 Physical Fact-Checking 11.2.2 Automatic Fact-Checking 11.3 Writing Mode–Based Approaches for Fake News Detection 11.3.1 Falsehood Identification and Analysis 11.3.2 Deception in NEWS 11.4 Dissemination-Based Approaches for Fake News Detection 11.4.1 Fake News Propagation Patterns 11.4.2 Models Based on Fake News Propagation 11.4.3 Dissemination-Based Fake News Detection 11.5 Integrity-Based Fake News Detection 11.5.1 Determining the Accuracy of Headline News 11.5.2 Determining the Accuracy of News Outlets 11.5.3 Determining the Accuracy of News Comments 11.5.4 Determining the Accuracy of Spreader News 11.6 Application of Machine Learning in Fake News Detection 11.6.1 Proposed Work 11.6.2 Proposed Methodology 11.6.3 Design and Workflow Diagram of Proposed System 11.6.4 Experimental Results and Analysis 11.7 Open Research Challenges in Fake News Detection 11.7.1 Redetection of Fake News 11.7.2 Identification of Trustworthy Contents 11.7.3 Fake News in Cross-Domain 11.7.4 Deep Learning and Fake News 11.7.5 Fake News and Social Media 11.8 Conclusion References 12. Authentication of Broadcast News on Social Media Using Machine Learning 12.1 Introduction 12.2 Literature Survey 12.3 Fake News Detection Using Machine Learning 12.3.1 Data Retrieval 12.3.2 Data Pre-Processing 12.3.3 Data Visualization 12.3.4 Tokenization 12.3.5 Feature Extraction 12.3.6 Machine Learning Algorithms for Fake New Detection 12.3.6.1 Support Vector Machine (SVM) 12.3.6.2 Naive Bayes Classifier 12.3.6.3 Decision Tree 12.3.6.4 ANN 12.3.7 Training and Testing Model 12.4 Comparative Study 12.5 Conclusion References 13. Application of Deep Learning in Facial Recognition 13.1 Introduction to Facial Recognition 13.2 Introduction to Deep Learning (DL) 13.3 Deep Learning Models for Facial Recognition 13.3.1 DeepFace 13.3.2 FaceNet 13.3.3 ArcFace 13.3.4 Baidu 13.3.5 Face Recognition Datasets 13.3.6 Comparison 13.3.7 Loss Functions Used to Improve the Network 13.3.7.1 Euclidean-Distance Based Loss 13.3.7.2 Softmax Loss and Its Variants 13.4 Scope and Challenges 13.5 Conclusion References 14. Application of Deep Learning in Deforestation Control and Prediction of Forest Fire Calamities 14.1 Introduction 14.2 Problems and Relevance to Today’s Society/Environmental Need 14.2.1 Deforestation 14.2.2 Animal, Insect Attack 14.2.3 National Security on Borders and Illegal Smuggling of Goods 14.2.4 Forest Fires 14.3 Brief Solution for the above Problems 14.4 Reasons and Causes of Wildfires 14.5 Methods to Detect, Predict and Control the Density of Forests 14.5.1 Methods to Detect Forest Fires 14.5.2 Methods to Predict Forest Fires 14.5.3 Deforestation Control 14.5.3.1 Dataset Required 14.5.3.2 Processing and Analysis of Images 14.5.3.3 Model Creation Using Logistic Regression 14.6 Conclusion Bibliography Journal Article News Article 15. Application of Convolutional Neural Network in Feather Classifications 15.1 Introduction 15.2 Multilayer Neural Networks 15.3 Convolutional Neural Networks 15.4 Greedy Snake Algorithm 15.5 Dataset 15.6 Proposed Methodology 15.6.1 Base Model 15.6.2 Data Augmentation 15.6.3 Implementation 15.7 Experimentation and Measures 15.7.1 K-Fold Validation 15.7.2 Measures 15.8 Results and Discussion 15.9 Conclusion 15.10 Future Scope References 16. Application of Deep Learning Coupled with Thermal Imaging in Detecting Water Stress in Plants 16.1 Introduction 16.2 Related Work 16.3 Proposed Methodology 16.3.1 Data Set 16.3.2 Data Augmentation 16.3.3 Pre-Processing of Thermal Images 16.3.4 Feature Extraction 16.4 Results 16.5 Conclusion References 17. Machine Learning Techniques to Classify Breast Cancer 17.1 Introduction 17.2 Literature Survey 17.3 Proposed Methodology 17.3.1 Dataset 17.3.2 Machine Learning Algorithms 17.3.2.1 Logistic Regression 17.3.2.2 Random Forest 17.3.2.3 Decision Trees 17.4 Implementation 17.4.1 Confusion Matrix 17.4.2 Performance Metrics 17.4.3 Comparison among Classification Algorithms 17.5 Results and Discussions 17.6 Conclusion References 18. Application of Deep Learning in Cartography Using UNet and Generative Adversarial Network 18.1 Introduction 18.2 Dataset and Preprocessing 18.2.1 Dataset 18.2.2 Data Preprocessing 18.3 Generative Models 18.3.1 UNet 18.3.2 Generative Adversarial Networks 18.3.2.1 BCELogItsLoss 18.3.3 Residual Neural Networks 18.4 Experimentation 18.4.1 Training of UNet 18.4.2 Training of a GAN 18.4.2.1 Training of Generator Module 18.4.2.2 Training of Discriminator Module 18.4.2.3 Combined Training 18.5 Results of the Experimentation 18.6 Future Applications 18.7 Conclusion References 19. Evaluation of Intrusion Detection System with Rule-Based Technique to Detect Malicious Web Spiders Using Machine Learning 19.1 Introduction 19.2 Intrusion Detection System 19.2.1 Network Intrusion Detection System 19.2.2 Host-Based Intrusion Detection Systems 19.3 Web Spider 19.3.1 Well-Behaved Web Spider 19.3.2 Malicious Web Spider 19.4 Proposed System 19.4.1 Weka 19.4.2 C4.5 Algorithm 19.4.3 J48 Classifier 19.4.4 Multilayer Perceptron Classifier 19.5 Evaluation of Proposed System Using Machine Learning Algorithm 19.5.1 IBk Classifier 19.5.2 Random Tree Classifier 19.6 Conclusion References 20. Application of Machine Learning to Improve Tourism Industry 20.1 Introduction 20.1.1 Types of Tourism 20.2 Industries Related to Tourism 20.2.1 Hotels 20.2.2 Restaurants 20.2.3 Retail and Shopping 20.2.4 Transportation 20.2.5 Travel Agencies 20.2.6 Tour Operator 20.2.7 Cultural Industries 20.3 Challenges of Tourism Industry 20.3.1 Taxation 20.3.2 Globalization 20.3.3 Marketing 20.3.4 Security 20.3.5 Terrorism 20.3.6 Infrastructure 20.3.7 Economy 20.3.8 Culture 20.3.9 Environment 20.3.10 Competition 20.3.11 Social Media 20.3.12 Pandemic 20.3.13 Price 20.4 Machine Learning and Deep Learning in Tourism 20.4.1 Machine Learning 20.4.1.1 Supervised Learning 20.4.1.2 Unsupervised Learning 20.4.1.3 Reinforcement Learning 20.4.2 Deep Learning 20.5 Sentiment Analysis 20.5.1 Block Diagram of Sentiment Analysis 20.5.2 Different Types of Sentiment Analysis 20.5.2.1 Fine-Grained 20.5.2.2 Emotion Detection 20.5.2.3 Aspect-Based 20.5.2.4 Intent Analysis 20.5.3 Sentiment Analysis Datasets 20.5.4 Sentiment Analysis Methods 20.5.4.1 Rule-Based Sentiment Analysis 20.5.4.2 Automated Sentiment Analysis 20.5.5 Sentiment Classification Techniques 20.5.6 Models for Sentiment Analysis 20.5.6.1 Logistic Regression 20.5.6.2 Random Forest 20.5.6.3 Support Vector Machine (SVM) 20.5.6.4 Naive Bayes Classifier 20.6 Sentiment Analysis-Based Research on Tourism 20.6.1 Sentiment Analysis Based on Environment 20.6.2 Sentiment Analysis Based on Locations and Hotels 20.6.3 Sentiment Analysis Based on Language 20.6.4 Sentiment Analysis Based on Trip Length 20.7 Summary References 21. Training Agents to Play 2D Games Using Reinforcement Learning 21.1 Introduction 21.1.1 Description 21.1.2 Problem Formulation 21.2 Review of Literature 21.3 Proposed Method/Design 21.3.1 Experimental Setup 21.3.2 Results and Discussion 21.3.3 Conclusion and Future Scope References 22. Analysis of the Effectiveness of the Non-Vaccine Countermeasures Taken by the Indian Government against COVID-19 and Forecasting Using Machine Learning and Deep Learning 22.1 Introduction 22.2 Related Work 22.3 Data Analysis 22.3.1 Top Ten States with Most Fatalities 22.3.2 Highest Fatality Rates 22.3.3 Population to Confirmed Cases Ratio 22.3.4 Analysis of Five-Phased Lockdown in India 22.4 ARIMA (Autoregressive Integrated Moving Average) 22.4.1 Mathematical Explanation of the ARIMA Model 22.4.2 Daily Death Count Forecasting 22.4.3 Daily Active Count Forecasting 22.4.4 Daily Confirmed Cases Count Forecasting 22.5 Recurrent Neural Networks and Long Short-Term Memory Networks 22.6 Conclusion Bibliography 23. Application of Deep Learning in Video Question Answering System 23.1 Introduction 23.2 Related Work 23.2.1 Visual Question Answering 23.2.1.1 Image Question Answering 23.2.1.2 VQA (Video Question Answering) 23.2.2 Emotion Detection 23.2.2.1 E.D. on Images 23.2.2.2 E.D. on Videos 23.2.3 Visual Captioning 23.2.3.1 Image Captioning 23.2.3.2 Video Captioning 23.2.4 Multitask Learning 23.3 AQAV 23.3.1 Overview 23.3.1.1 Emotion, Video and Question Embeddings 23.3.1.2 Vocabulary 23.3.1.3 Video Emotion Detector 23.3.2 Main VQA Route 23.3.2.1 Fusing Video, Emotion and Question Features 23.3.2.2 Feature Attention Techniques 23.3.3 Affective Route for Emotion Detection 23.3.3.1 Captioning-Module 23.3.3.2 Module–Text QA 23.3.4 Affective and Conventional Answers’ Prediction 23.4 Experiments and Results 23.4.1 Video Dataset 23.4.2 Experiment Setup and Results 23.4.2.1 Optimization 23.4.2.2 Performance Analysis 23.4.3 Comparing with Baseline Models 23.4.4 Validating Attention Model 23.4.5 Accuracy Analysis of AQAV Conventional Answers 23.4.6 Accuracy Analysis of AQAV Affective Answers 23.4.7 Qualitative Analysis 23.5 Conclusion References 24. Implementation and Analysis of Machine Learning and Deep Learning Algorithms 24.1 Introduction 24.2 Variable Terminology 24.3 Experimentation with Regularization Algorithms 24.3.1 Ridge Regularization 24.3.2 Lasso Regularization 24.3.3 Elastic Net Regularization 24.4 Decision Trees 24.4.1 CART (Classification and Regression Trees) 24.4.1.1 Regression Trees 24.4.1.2 Classification Trees 24.4.2 Iterative Dichotomiser (ID3) 24.4.3 C4.5 and C5.0 24.4.4 Decision Stumps 24.4.5 M5 Model Tree 24.5 Ensemble Learning 24.5.1 Bagging 24.5.1.1 Random Forest 24.5.2 Boosting 24.5.2.1 AdaBoost 24.5.2.2 Gradient Boosting 24.6 Implementation and Analysis of Deep Learning Algorithms 24.6.1 Feedforward Neural Networks 24.6.2 Convolutional Neural Networks 24.6.2.1 Convolution 24.6.2.2 Pooling 24.6.2.3 Results and Applications 24.6.3 Recurrent Neural Networks 24.6.4 Long Short-Term Memory Units 24.6.4.1 Applications and Advantages 24.6.5 Generative Adversarial Networks 24.6.6 Deep Belief Networks 24.7 Conclusion References 25. Comprehensive Study of Failed Machine Learning Applications Using a Novel 3C Approach 25.1 Introduction 25.2 Literature Review 25.3 3C Approach 25.4 Consolidation 25.5 Classification 25.6 Case Study and Failure Analysis 25.7 Results and Discussion 25.8 Conclusion 25.9 Future Scope References Index
Donate to keep this site alive
1. Disable the AdBlock plugin. Otherwise, you may not get any links.
2. Solve the CAPTCHA.
3. Click download link.
4. Lead to download server to download.