Intelligent Systems and Machine Learning for Industry: Advancements, Challenges, and Practices

Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Preface
About the Editors
Contributors
Chapter 1 A framework for a virtual reality-based medical support system
1.1 Introduction
1.2 Background
1.2.1 What is VR?
1.2.2 VR in healthcare education
1.2.3 Utilization of VR in the medical sector
1.3 Stakeholders of the system
1.3.1 Desirements of the stakeholders
1.4 System purpose and concept of operations
1.4.1 Concept of operations
1.4.1.1 Current and planned system
1.4.1.2 Functions of the system
1.4.1.3 Critical system requirements
1.4.2 Context diagram
1.4.2.1 First terminator
1.4.2.2 Second terminator
1.4.2.3 Third terminator
1.4.2.4 Fourth terminator
1.5 Systems requirements and use cases
1.5.1 Use cases
1.5.2 Block definition diagrams
1.6 Technical performance measure (TPM)
1.7 Conclusion
References
Chapter 2 ConvMax: Classification of COVID-19, pneumonia, and normal lungs from X-ray images using CNN with modified max-pooling layer
2.1 Introduction
2.2 Literature review
2.3 Proposed work
2.3.1 Proposed methodology
2.3.2 Dataset collection
2.3.3 Novel contributions of this study
2.3.4 System flow and related concepts
2.3.5 Proposed CNN architecture
2.4 Results and discussion
2.5 Conclusion
References
Chapter 3 Biorthogonal filter-based algorithm for denoising and segmentation of fundus images
3.1 Introduction
3.1.1 Motivation
3.1.2 Major contribution
3.1.3 Outcomes
3.1.4 Chapter organization
3.2 Literature review
3.2.1 Denoising of medical images
3.2.2 Segmentation techniques for medical images
3.3 Performance evaluation
3.3.1 Denoising performance metrics
3.3.2 Segmentation performance metrics
3.4 Biorthogonal wavelet transforms and filters
3.5 Investigational results
3.5.1 Noise suppression in images using the wavelet transform
3.5.2 Contrast enhancement of the images for improved segmentation
3.5.3 Findings
3.6 Discussion and conclusions
References
Chapter 4 Deep learning-based automatic detection of breast lesions on ultrasound images
4.1 Introduction
4.2 Methodology
4.2.1 Block diagram of proposed method
4.2.2 Breast ultrasound dataset
4.2.3 Preprocessing
4.2.3.1 Overview
4.2.3.2 Block diagram of the proposed speckle reduction method
4.2.3.3 Speckle reduction by circular hybrid median filter technique
4.2.3.4 Algorithm
4.2.3.5 Performance indices
4.2.4 Segmentation
4.2.4.1 Need for image segmentation
4.2.4.2 CNN-based image segmentation
4.2.4.3 Residual network
4.2.4.4 Implementation
4.2.4.5 Analyzing the network
4.2.4.6 Performance indices
4.3 Results and discussions
4.3.1 Preprocessing results
4.3.2 Segmentation results
4.4 Conclusion
References
Chapter 5 Heart disease prediction using enhanced machine learning techniques
5.1 Introduction
5.2 Overview of machine learning techniques
5.3 Heart disease datasets
5.4 Notable heart disease prediction studies
5.5 Proposed heart disease prediction approach
5.5.1 Decision tree
5.5.2 Logistic regression
5.5.3 Support vector machine (SVM)
5.5.4 Random forest
5.5.5 XGBoost
5.5.6 Adaptive boosting
5.6 Experimental results and discussion
5.7 Conclusion
References
Chapter 6 Immersive technologies in healthcare education
6.1 Introduction
6.2 Background
6.2.1 Research questions
6.2.2 Reviews of VR/AR/MR/XR applications in the healthcare domain
6.3 Immersive technologies
6.3.1 Virtual reality (VR)
6.3.2 Augmented reality (AR)
6.3.3 Mixed reality (MR)
6.3.4 Extended reality (XR)
6.4 Immersive technologies in education
6.4.1 Research gaps
6.5 Conclusion
References
Chapter 7 Implications of technological trends toward smart farming
7.1 Introduction
7.1.1 Background
7.1.2 Motivation
7.1.3 Major contribution
7.1.4 Paper organization
7.2 What smart agriculture can do
7.3 IoT in smart agriculture
7.3.1 Major equipment and technologies
7.4 Intelligent systems in smart agriculture
7.4.1 Subsets of intelligent systems
7.4.1.1 Why artificial neural networks?
7.4.1.2 What distinguishes deep learning from machine learning techniques?
7.5 Smart measurements
7.5.1 Spectroscopic techniques
7.5.2 Image processing
7.5.3 Electronic nose in food quality
7.5.3.1 How electronic nose (e-nose) works
7.5.3.2 e-Nose sensing system
7.5.3.3 Pattern recognition algorithms
7.6 IoT integration with intelligent systems in smart agriculture
7.7 Summary
References
Chapter 8 A smart sensing technology for monitoring marine environment conditions
8.1 Introduction
8.2 Overview of IoT
8.2.1 Perception and execution
8.2.2 Data transmission
8.2.3 Preprocessing data
8.2.4 Application
8.2.5 The organizational layer
8.3 Characteristics of sensors
8.4 Platform for sensors
8.4.1 Marine observation system with static point
8.4.2 Subsurface floats
8.4.3 Remotely autonomous and operated vehicles (ROV)
8.4.4 Satellite
8.4.5 Remotely piloted aircraft (RPA)
8.5 Type of sensors for data collection
8.5.1 Acoustics
8.5.2 Cameras
8.5.3 Satellite sensors
8.5.4 Biosensors
8.6 Challenges in application
8.7 Conclusion
References
Chapter 9 Managing agriculture pollution using life-cycle assessment and artificial intelligence methods
9.1 Introduction
9.2 Benefits of artificial intelligence in agriculture
9.2.1 IoT drives data analytics
9.2.2 Drone-assistant technology
9.2.3 Weather prediction
9.2.4 Monitoring soil and crop health
9.2.5 Artificial intelligence can assist with labor shortages
9.2.6 Farm data analysis and pest monitoring
9.3 Managing environmental pollution in the agricultural sector
9.4 Advantages of LCA–AI integration
9.5 Conclusion
References
Chapter 10 Ensemble techniques for effective prediction of crop selection in the Coastal Andhra deltaic region
10.1 Introduction
10.1.1 Agriculture in India
10.1.2 Agriculture in Andhra Pradesh
10.1.3 Essential nutrients for production of crops
10.1.4 Challenges in agriculture
10.1.5 Data mining techniques in agriculture
10.1.6 Organization of the chapter
10.2 Related work
10.3 Learning methods
10.3.1 Machine learning
10.3.1.1 Classification tree
10.3.1.2 K-nearest neighbor algorithm
10.3.2 Ensemble methods
10.3.2.1 Random space
10.3.2.2 Bagging
10.3.2.3 AdaBoosting
10.4 Experimental analysis
10.4.1 Dataset
10.4.2 Process flow
10.4.2.1 Step 1: Selection of suitable dataset for crop prediction
10.4.2.2 Step 2: Identification of training data and testing data
10.4.2.3 Step 3: Build the crop advisor models
10.4.2.4 Step 4: Test the models
10.4.2.5 Step 5: Comparison of models
10.4.3 Performance metrics
10.4.3.1 Accuracy
10.4.3.2 Precision
10.4.3.3 Recall
10.4.3.4 F1-score
10.4.4 Experimentation
10.5 Results and discussion
10.6 Conclusion
References
Chapter 11 Artificial intelligence-based quality inference for food processing industry applications
11.1 Introduction
11.2 Food quality and safety
11.2.1 Artificial intelligence in food quality and safety
11.3 Non-destructive techniques
11.3.1 Near infrared
11.3.2 Hyperspectral imaging
11.3.3 Thermal imaging
11.3.4 e-Nose and e-tongue
11.4 Conclusion
References
Chapter 12 A study on intelligent systems and their influence on smarter defense service
12.1 Introduction
12.2 Artificial intelligence and its current status
12.2.1 AI as a growing technology
12.2.2 AI-built technologies
12.2.3 Fear of AI
12.3 Artificial intelligence and its usage in defense
12.3.1 Training
12.3.2 Surveillance
12.3.3 Artillery
12.3.4 Cyberattacks
12.3.5 Cognitive radio and cognitive electronic warfare
12.3.6 Computational military reasoning (tactical artificial intelligence)
12.3.7 Intelligent and autonomous unmanned weapon systems
12.3.8 Information processing, intelligent analysis, and data fusion using AI
12.4 Practical use of AI in military applications
12.4.1 Application of neural networks in object location
12.4.2 Location of underwater mines using deep convolution neural network
12.4.3 Application of neural networks in cybersecurity
12.5 Challenges of using AI in defense operations
12.6 Conclusion
References
Chapter 13 Steam turbine controller using fuzzy logic
13.1 Introduction
13.1.1 Description of technology at block level
13.1.2 Practical realization of fuzzy controller
13.2 Description of methodology used for implementation
13.2.1 Choosing fuzzy controller inputs and outputs
13.2.2 Linguistic descriptions
13.2.3 Rules
13.2.4 Rule-bases
13.2.5 Operations
13.2.6 Fuzzy quantification of knowledge
13.2.7 Defuzzification methods
13.3 Fuzzy logic controller
13.3.1 Problems encountered and their solutions
13.3.2 Deciding the input variables
13.3.3 Deciding membership functions
13.3.4 Deciding range
13.4 Deciding the type of function
13.5 Results and discussions
13.6 Concluding remarks
Conflict of interest
References
Chapter 14 Speech recognition for Indian-accent English using a transformer model
14.1 Introduction
14.2 Problem statement
14.3 Proposed methodology
14.4 Data overview
14.4.1 Text files
14.4.2 Data preprocessing
14.5 Technologies used
14.5.1 Transformers
14.5.2 Attention
14.5.3 Elliptic curve cryptography (ECC)
14.5.4 Keys in ECC
14.5.5 Addition in ECC
14.5.6 Multiplication in ECC
14.6 Literature survey
14.6.1 “SpecAugment: a simple data augmentation method for automatic speech recognition”
14.6.1.1 Augmentation policy
14.6.1.2 Learning rate schedules
14.6.1.3 Results
14.6.2 “Speech recognition using deep neural networks: a systematic review”
14.6.2.1 Machine learning techniques
14.6.2.2 Generative models
14.6.2.3 Deep neural networks
14.6.2.4 Conclusion
14.6.3 “Deep learning: from speech recognition to language and multimodal processing”
14.6.3.1 Introduction
14.6.3.2 From deep generative models to DL models
14.6.3.3 Advanced architectures
14.6.3.4 Summary
14.6.4 “Speech commands: a dataset for limited vocabulary-speech recognition”
14.6.4.1 Abstract
14.6.4.2 Introduction
14.6.4.3 Conclusion
14.6.5 “A neural attention model for speech command recognition”
14.6.5.1 Abstract
14.6.5.2 Introduction
14.6.5.3 Neural network implementation
14.6.5.4 Conclusion
14.6.6 “Automatic speech recognition using different neural network architectures – a survey”
14.6.6.1 Convolutional neural network (CNN)
14.6.6.2 Recurrent neural network (RNN)
14.6.7 Towards end-to-end speech recognition with recurrent neural networks
14.7 Implementation
14.7.1 Input audio processing
14.7.2 Encoder
14.7.3 Target sentence encoding
14.7.4 Decoder
14.8 Results and conclusions
References
Chapter 15 Stock market prediction using sentiment analysis with LSTM and RFR
15.1 Introduction
15.2 Background study
15.2.1 Literature review
15.2.2 Time series
15.2.3 Deep learning – LSTM
15.2.4 Random forest
15.2.5 Sentiment analysis
15.3 Methodology
15.3.1 Data source
15.3.2 Evaluation criteria
15.4 Approach and implementation
15.4.1 Data preprocessing
15.4.2 Visualization
15.4.3 Time series analysis
15.4.4 Sentiment analysis
15.4.5 LSTM
15.4.6 RFR
15.5 Results
15.6 Conclusion
References
Chapter 16 A systematic and exhaustive analysis of intelligent software effort estimation models
16.1 Introduction
16.1.1 Algorithmic models
16.1.2 Expert systems
16.1.3 Soft computing techniques
16.1.4 Major contributions
16.1.5 Organization of the chapter
16.2 Literature survey
16.3 Research outcomes in software effort estimation
16.3.1 Scopus database search
16.3.2 Initial search outcomes
16.4 Systematic analysis
16.4.1 Statistical analysis
16.4.2 Network analysis
16.4.2.1 Citation analysis of documents
16.4.2.2 Citation analysis of sources
16.4.2.3 Citation analysis by authors
16.4.2.4 Analysis of citations by organization
16.4.2.5 Citation analysis by country
16.4.2.6 Co-citation analysis by cited references
16.4.2.7 Co-citation analysis by cited sources
16.4.2.8 Co-citation analysis by cited authors
16.5 Results and discussions
16.6 Research directions
16.7 Conclusion
References
Index