Introduction
Contents
Part I Fundamentals of Industry 4.0
1 Opting for Industry 4.0: Challenge or Opportunity
1.1 Introduction
1.1.1 Role of Technologies in Industry 4.0 Transformation
1.1.2 Key Technologies to Transform Production Industry
1.2 Challenges When One Wants to Switch to Industry 4.0
1.2.1 Major Challenges
1.2.2 Some More Challenges
1.3 Benefits of Industry 4.0
1.4 Applications
1.5 Societal Impact
1.5.1 Relation Between Profit and Purpose
1.5.2 Employee and Customer Advocacy Is Increased
1.6 Case Study: Challenges in Manufacturing Sector
1.7 Conclusion
References
2 Exploring Human Computer Interaction in Industry 4.0
2.1 Introduction
2.2 Related Work
2.3 Research Questions
2.4 Research Solutions
2.5 Discussion and Recommendations
2.6 Conclusion and Future Work
References
3 Embedding Affect Awareness in e-Learning: A Systematic Outline of the Literature
3.1 Introduction
3.2 Motivation
3.3 Research Strategy and Research Questions
3.4 Literature Review
3.5 Survey Outcome and Discussion
3.6 Conclusion
References
4 Edge Computing: A Paradigm Shift for Delay-Sensitive AI Application
4.1 Introduction
4.2 Edge Computing Overview
4.2.1 The Origin of Edge Computing
4.2.1.1 Three Paradigms of Edge Computing: Cloudlets, Fog Computing, and Mobile Edge Computing
4.2.2 Criteria-Wise Difference Between Edge Computing and Cloud Computing
4.2.3 Layered Architecture of Edge Computing
4.2.3.1 Device Layer
4.2.3.2 The Edge Layer
4.2.3.3 The Cloud Layer
4.2.4 Software and Hardware Requirements to Implement Edge Computing (Table 4.2)
4.2.5 Characteristics of Edge Computing
4.2.5.1 Edge Computing Has Many Characteristics [9] as Cloud Computing
4.2.5.2 Close Proximity to the End Device
4.2.5.3 Support for Mobility Management
4.2.5.4 Location Awareness
4.2.5.5 Low Latency
4.2.5.6 Low Computation Power
4.2.6 Disadvantages of Edge Computing
4.2.7 Overview of Edge AI
4.2.8 Why Deep Learning with Edge Computing
4.2.9 Edge Intelligence-Enabled Applications of IoT
4.2.9.1 Smart Wearables
4.2.9.2 Smart City
4.2.9.3 Smart Home
4.2.9.4 Smart Building
4.2.9.5 Smart Grid
4.2.9.6 Smart Vehicle
4.2.9.7 Smart Multimedia
4.2.9.8 Video Analytics
4.2.9.9 Adaptive Video Streaming
4.2.9.10 Smart Transportation
4.2.9.11 Autonomous Driving
4.2.9.12 Traffic Analysis
4.2.9.13 Traffic Signal Control
4.2.10 Challenges in Edge-Enabled IoT Systems
4.2.10.1 Modal Training
4.2.10.2 Modal Deployment
4.2.10.3 Delay-Sensitive Applications
4.2.10.4 Hardware and Software Support
4.2.10.5 Integration and Heterogeneity
4.2.10.6 Naming
4.2.11 Research Opportunities in Edge AI and or Edge Computing
4.2.12 Conclusion
References
Part II Emerging Trends in Artificial Intelligence
5 CBT-Driven Chatbot with Seq-to-Seq Model for IndianLanguages
5.1 Introduction
5.2 Literature Survey
5.3 Proposed Work
5.3.1 Flow of the System
5.3.2 Architecture
5.4 Implementation
5.5 Results and Discussion
5.5.1 Training Dataset
5.5.2 Application Overview and the Chatbot Interface Designed
5.6 Conclusion
References
6 A Review of Predictive Maintenance of Bearing Failures in Rotary Machines by Predictive Analytics Using Machine-Learning Techniques
6.1 Introduction
6.2 Survey and Analysis for Related Work
6.3 Technical Background
6.4 Predictive Maintenance and Machine-Learning Techniques
6.4.1 Supervised Learning
6.4.1.1 Classification
6.4.1.2 Regression
6.4.2 Unsupervised Learning
6.4.3 Reinforcement Learning
6.5 Challenges
6.6 Applications of ML Algorithms in PdM
6.7 Discussion and Conclusions
References
7 Crop and Fertilizer Recommendation System Using Machine Learning
7.1 Introduction
7.2 Literature Survey
7.3 Proposed System
7.4 Implementation and Results
7.5 Crop Recommendation Methodology
7.6 Fertilizer Recommendation Methodology
7.7 Conclusion
References
8 Comparative Analysis of Machine Learning Algorithms for Intrusion Detection System
8.1 Introduction
8.2 Related Work
8.3 Methodology (Fig. 8.1)
8.3.1 Dataset
8.3.2 Binary Classifiers
8.3.2.1 Random Forest Classifier
8.3.2.2 AdaBoost Classifier
8.3.2.3 Logistic Regression Classifier
8.3.2.4 Linear Support Vector Machine
8.3.3 One-Class Classifiers
8.3.3.1 OneClass SVM
8.3.3.2 Isolation Forest
8.3.4 Autoencoders
8.4 Results and Discussion
8.5 Conclusion
References
9 Facial Recognition System Using Transfer Learning with the Help of VGG16
9.1 Introduction
9.2 Literature Review
9.3 Proposed Methodology
9.3.1 Convolutional Neural Network (CNN)
9.3.1.1 Convolution Layer
9.3.1.2 Pooling Layer
9.3.1.3 ReLU Correction Layer
9.3.1.4 Fully Connected (FC) Layer
9.3.2 VGG-16 Neural Network Model
9.3.3 Transfer Learning
9.3.3.1 Dataset Collection and Cleaning
9.3.3.2 Loading the VGG16 Model and Fine-Tuning
9.3.4 Loading Dataset and Training Model
9.3.4.1 Validation and Prediction (Fig. 9.11)
9.3.4.2 Output (Fig. 9.12)
9.4 Result and Discussion
9.5 Future Work
References
10 Digitization in Teaching and Learning: Opportunities and Challenges
10.1 Introduction
10.2 Paper Organization
10.3 Related Work
10.4 Proposed Methodology
10.5 Dataset and Data Description
10.6 Results and Discussion
10.6.1 Department
10.6.2 Technology
10.6.3 Participation
10.6.4 Time
10.6.5 Online Practical
10.6.6 Teacher-Student Interaction
10.6.7 Technology
10.6.8 Focus
10.6.9 Internet Connectivity
10.6.10 Exam
10.6.11 Area of Living
10.7 Results from Teacher's Survey
10.8 Conclusion
References
Part III AI Based Data Management, Architecture and Frameworks
11 AI-Based Autonomous Voice-Enabled Robot with Real-Time Object Detection and Collision Avoidance Using Arduino
11.1 Introduction
11.2 Literature Survey
11.3 Proposed System Design and Methodology
11.3.1 Flowchart (Fig. 11.2)
11.3.2 System Requirements
11.3.2.1 Arduino UNO (Fig. 11.3)
11.3.2.2 L298N Motor Driver (Fig. 11.4)
11.3.2.3 Bluetooth Module HC05 (Fig. 11.5)
11.3.2.4 Ultrasonic Sensor (Fig. 11.6)
11.3.2.5 BO Motors
11.3.2.6 Connecting Wires
11.3.2.7 ESP32 Camera Module (Fig. 11.7)
11.3.2.8 Power Supply
11.3.2.9 Wheels
11.3.2.10 Servo Motor (Fig. 11.8)
11.3.2.11 USB to TTL Module (Fig. 11.9)
11.3.3 Project Methodology
11.3.4 Voice-Controlled System
11.3.5 Algorithm
11.4 Robot Implementation
11.4.1 Libraries Used
11.4.2 Android Application Design (Fig. 11.11)
11.4.3 Development Software
11.4.4 Hardware Implementation (Fig. 11.14)
11.5 Results and Discussion
11.6 Conclusion and Discussion
11.7 Future Scope
References
12 Real-Time Interactive AR for Cognitive Learning
12.1 Introduction
12.2 Related Work
12.3 Need and Motivation
12.4 Proposed Work
12.4.1 Scalable Cloud Integration
12.4.2 Input Interface
12.4.3 Computational Engines
12.4.4 Language Processing Engine
12.4.5 Knowledge Processing
12.4.6 Output Interface
12.5 Result and Future Discussion
12.6 Conclusion
References
13 Study and Empirical Analysis of Sentiment Analysis Approaches
13.1 Introduction
13.2 Literature Survey
13.2.1 Lexicon-Based Methods
13.2.2 Machine-Learning-Based Methods
13.2.3 Deep-Learning-Based Methods
13.2.4 Datasets
13.3 Experiment Methodology
13.3.1 Pre-processing
13.3.1.1 Pre-processing on Large Movie Reviews Dataset
13.3.1.2 Pre-processing on Sentiment140 Dataset
13.3.1.3 Pre-processing on Amazon Baby Dataset
13.3.2 Methods
13.3.3 Evaluation Metrics
13.4 Results
13.4.1 Results of Lexicon-Based Methods
13.4.2 Results of Machine-Learning-Based Methods
13.4.3 Results of Deep-Learning-Based Methods
13.5 Conclusion
13.6 Future Work
References
14 Sign Language Machine Translation Systems: A Review
14.1 Introduction
14.2 Sign Language
14.3 Challenges of Sign Language Machine Translation
14.3.1 Simultaneity in Articulation
14.3.2 Non-manual Features
14.3.3 Signing Space
14.3.4 Morphological Incorporation
14.4 Sign Language Writing/Representation Systems
14.4.1 Annotation Systems
14.4.2 Pictorial Systems
14.4.3 Symbolic Systems
14.5 Overview of Sign Language Machine Translation at the Global Level
14.5.1 The Zardoz System
14.5.2 Translation from English to American Sign Language by Machine (TEAM)
14.5.3 Visual Sign Language Broadcasting (ViSiCast)
14.5.4 A Multi-path Architecture
14.5.5 Research by RWTH Aachen Group
14.5.6 Project Web-Sign
14.5.7 Machine Translation Using Examples (MaTrEx)
14.5.8 Japanese to Japanese Sign Language (JSL) Glosses Using a Pre-trained Model
14.5.9 Sign Language Production Using Generative Adversarial Networks
14.6 Overview of Sign Language Machine Translation for Indian Languages
14.6.1 INGIT: Limited Domain Formulaic Translation from Hindi Strings to Indian Sign Language
14.6.2 Dictionary-Based Translation Tool for Indian Sign Language
14.6.3 Indian Sign Language Corpus for the Domain of Disaster Management
14.6.4 Indian Sign Language from Text
14.7 Gap Analysis
14.8 Discussion on the Designed Prototype and Proposed Enhancement
14.9 Conclusion
14.10 Future Scope
References
15 Devanagari Handwritten Character Recognition Using Dynamic Routing Algorithm
15.1 Introduction
15.2 Literature Review
15.3 Problem with CNN
15.4 Capsule Network
15.5 Dynamic Routing Between Capsules
15.6 Introduction of Devanagari Character Set
15.7 Challenges in Recognizing Devanagari Character
15.8 Experiment
15.9 Results and Discussion
15.10 Conclusion and Future Scope
References
Part IV Security for Industry 4.0
16 Predictive Model of Personalized Recommender System of Users Purchase
16.1 Introduction
16.2 Related Work
16.3 Gap Analysis
16.4 Research Hypotheses
16.4.1 Personalized Recommendation, Privacy Concerns
16.4.2 Personalized Recommendation and Satisfaction
16.4.3 Privacy Concern with Trust
16.4.4 Privacy Concerns and Purchase Intention
16.4.5 Satisfaction, Trust, and Purchase Intention
16.5 Methodology of Research
16.5.1 Data Collection and Sampling
16.5.2 Data Analysis and Measurement Model
16.5.3 Confirmatory Factor Analysis and Validity Test
16.5.4 Structural Equation Modeling (SEM)
16.6 Result and Discussion
16.7 Conclusions and Future Scope of Research
References
17 Rethinking Blockchain and Machine Learning for Resource-Constrained WSN
17.1 Introduction
17.2 Related Work
17.2.1 Conventional Trustworthy Routing Systems
17.2.2 Blockchain Network-Based Routing Mechanisms
17.3 Routing Approaches with Reinforcement Learning Algorithms
17.4 Blockchain and Reinforcement Learning Mechanisms to Improve Communication Network Routing Security and Efficiency in WSNs
17.5 Blockchain Technology
17.6 Routing Algorithm Based on Reinforcement Learning and Blockchain
17.7 Blockchain Network Procedure
17.8 Conclusions
References
18 Secure Data Hiding in Binary Images Using Run-Length Pairs
18.1 Introduction
18.2 Literature Survey
18.3 Gap Analysis
18.4 Proposed Technique
18.4.1 Compressed Data Preparation
18.4.2 Information Hiding Process
18.4.3 Information Extraction
18.5 Algorithms
18.5.1 Arithmetic Coding Algorithm
18.5.2 Embedding Algorithm
18.5.3 Extraction Algorithm
18.6 Result Analysis
18.7 Discussion
18.8 Conclusion
References
19 Privacy-Enhancing Techniques for Gradients in Federated Machine Learning
19.1 Introduction
19.2 Literature Review
19.3 FL Architecture
19.4 Experiments and Results
19.5 Privacy-Enhancing Techniques for FL
19.5.1 Secure Multiparty Computation (SMC)
19.5.2 Differential Privacy Preservation (DPP)
19.5.3 Homomorphic Encryption (HE)
19.5.4 Trusted Execution Environments (TEE)
19.6 Conclusion and Future Scope
References
Part V Software Language Implementation, Linguistics, and Virtual Machines
20 Multi-component Interoperability and Virtual Machines: Examples from Architecture, Engineering, Cyber-Physical Networks, and Geographic Information Systems
20.1 Introduction
20.2 Hypergraph Data Modeling
20.2.1 Hypergraphs as General-Purpose Data Models
20.2.2 Examples: Building Information Management and Medical Imaging
20.2.3 Virtual Machines in the Context of Data Metamodels and Database Engineering
20.2.4 Database Engineering and Type Theory
20.3 GIS Databases and Digital Cartography
20.3.1 Geospatial Data and GUI Events
20.3.2 Representing Functional Organization
20.4 Conclusion
References
21 Virtual Machines and Hypergraph Data/Code Models: Graph-Theoretic Representations of Lambda-Style Calculi
21.1 Introduction
21.2 Virtual Machines and Hypergraph Code Models
21.2.1 Applicative Structures and Mathematical Foundations
21.2.2 Hypergraph Models of Calling Conventions
21.3 Semantic Interpretation of Syntagmatic Graphs
21.3.1 Distinguishing Non-constructive from Extensional Type Semantics
21.3.2 Syntagmatic Graph Sequences as a Virtual Machine Protocol
21.4 Conclusion
References
22 GUI Integration and Virtual Machine Constructions for Image Processing: Phenomenological and Database Engineering Insights into Computer Vision
22.1 Introduction
22.2 Type-Theoretic Constructions at the Virtual Machine Level
22.2.1 Issues with Overflow/Underflow and Loop Termination
22.2.2 Different Variations on Enumeration Types
22.3 Integrating Virtual Machines with Image-Processing Operations
22.3.1 Exposing GUI Functionality
22.3.2 Extending Host Applications with Image-Processing Workflows
22.3.3 Manhattan/Chebyshev Distances and ``BlackβGrey'' Grids
22.3.4 XCSD Operators as Representative Image-Processing Functions
22.4 An Example Image-Processing Pipeline
22.4.1 From Keypoints to Superpixels
22.4.2 Interactive Workflows and Assessments
22.5 Conclusion
References
23 The Missing Links Between Computer and Human Languages: Animal Cognition and Robotics
23.1 Introduction
23.1.1 Comments on Methodology
23.2 Animal Cognition and Talking Dogs
23.2.1 Lessons for Natural Language
23.3 Joint Attention and the Foundations of Language
23.3.1 Learning from Humans
23.4 Conclusion
23.4.1 Robotics and Environment Models
References
24 GUIs, Robots, and Language: Toward a Neo-Davidsonian Procedural Semantics
24.1 Introduction
24.2 Semantics and Situational Change
24.2.1 The (Provisional) Semantics of Syntactic Dis-ambiguation
24.2.2 From Natural to Computer Languages
24.3 GUIs, Robots, and Environments
24.3.1 The Semantics of GUI Control State
24.3.1.1 Extending Object Orientation
24.3.2 3D Graphics and Robotics Front Ends
24.4 Conclusion
References
Index