Smart Computing and Self-Adaptive Systems (Computational Intelligence Techniques)

Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Editors
Contributors
Chapter 1: Using Luong and Bahdanau Attention Mechanism on the Long Short-Term Memory Networks: A COVID-19 Impact Prediction Case Study
	1.1 Introduction: Deep Learning
	1.2 RNN and LSTMs
		1.2.1 Recurrent Neural Networks
		1.2.2 LSTM Architecture
			1.2.2.1 Input Gate
			1.2.2.2 Forget Gate
			1.2.2.3 Cell State
			1.2.2.4 Output Gate
	1.3 Sequence Prediction
	1.4 Attention Mechanism
		1.4.1 Bahdanau Attention Mechanism
	1.5 Luong Attention Mechanism
		1.5.1 Comparison of Luong and Bahdanau Attention Mechanism
	1.6 COVID-19 Data Prediction
	1.7 LSTM Implementation Using Keras
		1.7.1 LSTM Sequence-to-Sequence Model
		1.7.2 LSTM Sequence-to-Sequence Implementation with Luong Attention
		1.7.3 LSTM Sequence-to-Sequence with Bahdanau Attention
	1.8 Results and Discussion
	1.9 Conclusion
	References
Chapter 2: A Novel Missing Data Imputation Algorithm for Deep Learning-Based Anomaly Detection System in IIoT Networks
	2.1 Introduction
	2.2 An Overview on Existing ADS Systems
	2.3 Proposed ADS System
		2.3.1 Overview of the Proposed EDMDI Algorithm-Based ADS System
		2.3.2 Architecture of the Proposed Scheme
		2.3.3 Enhanced DNN for Missing Data Imputation Algorithm
	2.4 Theoretical Analysis
	2.5 Results and Discussion
		2.5.1 Datasets
		2.5.2 Evaluation Metrics
		2.5.3 Experimental Results
		2.5.4 Performance Comparison of DNN-Based ADS System and EDMDI Algorithm-Based ADS System
	2.6 Conclusion
	References
Chapter 3: Internet of Things: Concept, Implementations and Applications
	3.1 Introduction
	3.2 Literature Review
	3.3 Sensors and Actuators
	3.4 Introduction to Arduino
		3.4.1 Size of the Board
	3.5 Concept of Smart Cities and Smart Homes
		3.5.1 Smart Home
		3.5.2 Smart Cities
	3.6 Introduction to IIoT
	3.7 Implementation of IoT in Various Fields
	3.8 Challenges in IoT
		3.8.1 Privacy Concern
		3.8.2 Complexity
		3.8.3 Dependency
		3.8.4 Society
	3.9 Conclusion and Future Aspect
	References
Chapter 4: Output-Oriented Multi-Pane Mail Booster: Data Crawling and Results in All Category Panes of a Mail
	4.1 Introduction
	4.2 Related Work
	4.3 Proposed Methodology
	4.4 Results
	4.5 Conclusion
	References
Chapter 5: Eyesight Test through Remote Virtual Doctor Using IoT
	5.1 Introduction
	5.2 Related Work
	5.3 Proposed Work
	5.4 Results
	5.5 Conclusion
	References
Chapter 6: Recent Trends and Advances in Deep Learning-Based Internet of Things
	6.1 Introduction
	6.2 Advanced Deep Learning Techniques
		6.2.1 Unsupervised and Transmit Learning
		6.2.2 Online Learning
		6.2.3 Optimization Techniques in Deep Learning
		6.2.4 Deep Learning in Distributed Systems
	6.3 Applications of IoT Using Deep Learning
		6.3.1 IoT Security
		6.3.2 Smart Healthcare
		6.3.3 Smart Home
		6.3.4 Smart Conveyance
		6.3.5 Smart Industry
		6.3.6 Smart Agriculture
	6.4 Conclusion and Future Aspects
	References
Chapter 7: Prediction and Classification Analysis of Type-2 Diabetes Using Machine Learning Approaches
	7.1 Introduction
	7.2 Background
	7.3 The Different Prediction Models of Machine Learning
		7.3.1 SVM
		7.3.2 Naïve Bayes
		7.3.3 Logistic Regression
		7.3.4 K-NN
		7.3.5 Random Forest
	7.4 Methodology and Model Diagram of the Proposed Work
		7.4.1 Describes the Methodology and Model Diagram for the Proposed Work
	7.5 Results and Their Experiments
	7.6 Conclusion and Future Scope
	References
Chapter 8: Internet of Thing-Based Monitoring Systems and Their Applications
	8.1 Introduction
	8.2 Advantages
	8.3 Applications in IoT
	8.4 General Architecture of the System
	8.5 Components Used by IoT
	8.6 Raspberry Pi/Arduino
	8.7 RFID Tag
	8.8 Condition Relay
	8.9 Bluetooth/Wi-Fi
	8.10 Data Server
	8.11 Data Analysis: ThingSpeak
		8.11.1 Features of ThingSpeak
	8.12 Sensors
	8.13 Types of Sensors
		8.13.1 PIR Sensor
		8.13.2 Temperature Sensor
		8.13.3 IR Sensor
		8.13.4 UV Sensor
		8.13.5 Touch Sensor
	8.14 IoT-Based Monitoring System Applications
		8.14.1 Health Monitoring System
		8.14.2 Temperature Monitoring System
		8.14.3 Smart Irrigation System for Agriculture
	8.15 Machine Monitoring System
	8.16 Marine Radioactivity Monitoring System
	8.17 Vehicle Tracking and Fuel Monitoring System
	8.18 Android-Based Monitoring System
	8.19 Applied Algorithm
		8.19.1 Performance of Symmetric Key
		8.19.2 Blowfish Algorithm
		8.19.3 Cipher-Attribute-Based Encryption
	8.20 Conclusion
	References
Chapter 9: A Progressive Method to Monitor Power Using IoT
	9.1 Introduction
	9.2 Literature Study
		9.2.1 System Design to Detect Load Automatically
			9.2.1.1 Collection of Data on Power Consumption
			9.2.1.2 Extracting Data of Power Load
			9.2.1.3 Processing of Power Load Data
			9.2.1.4 Tracking of Data
	9.3 Objectives of the Power Monitoring System
	9.4 Existing System
	9.5 Power Monitoring
	9.6 Internet of Things (IoT)
		9.6.1 Applications of IoT
			9.6.1.1 Big Data Analytics
				9.6.1.1.1 Xplenty
				9.6.1.1.2 Power BI
				9.6.1.1.3 Microsoft HDInsight
				9.6.1.1.4 Skytree
				9.6.1.1.5 Talend
				9.6.1.1.6 Splice Machine
				9.6.1.1.7 Spark
				9.6.1.1.8 Plotly
				9.6.1.1.9 Apache SAMOA
				9.6.1.1.10 Lumify
				9.6.1.1.11 Elastic Search
				9.6.1.1.12 R Programming
	9.7 Artificial Intelligence
	9.8 Cloud Storage
		9.8.1 Ubidots Cloud Service
			9.8.1.1 Ease of Use
			9.8.1.2 Support
			9.8.1.3 Fairness
			9.8.1.4 Sensors
	9.9 Architecture for the Proposed Power Monitoring System
	9.10 Sensor Components Used for Power Monitoring (Figures 9.2–9.5)
	9.11 Future Bill Prediction
	9.12 Conclusion
	References
Chapter 10: Cognitive Computing-Powered, NLP-Based, Autonomous Question-Answering System
	10.1 Introduction
	10.2 Background
	10.3 Concept Modeling of “Concept Model”, “Experience/Knowledge-Base Model”, and the “Interface Model” of the Proposed Cognitive Agent “QAEdu”
		10.3.1 Proposed Architecture
		10.3.2 Question Understanding
			10.3.2.1 Evaluation of Extraction Process for Qtype and Its Semantic Components
		10.3.3 Linguistic Component Extraction (Entity Disambiguation and Association Detection) from the Question
	10.4 Implementation of Experience and Knowledge Model
		10.4.1 Knowledge-Base Construction
			10.4.1.1 Evaluation of Extraction Process of Triples from the Documents for the Construction of Knowledge Base
		10.4.2 Methods for Answer Searching and Extraction from the Knowledge Base
	10.5 Methodology
	10.6 Implementation of Interface Model
		10.6.1 Experimental Design and Result
		10.6.2 Interface Module Designing
		10.6.3 Result
		10.6.4 Evaluation of QAEdu Question-Answering System
	10.7 Conclusion
	10.8 Challenges and Future Aspects
		10.8.1 Hardware Implementation for Providing the Embodiment to the Cognitive Agent
	10.9 Discussion
	References
Chapter 11: Smart Cities and Industry 4.0
	11.1 Introduction
		11.1.1 Introduction to Smart City 4.0
		11.1.2 Smart City vs. Industry
	11.2 4IR
		11.2.1 Four Concepts on Industry 4.0
			11.2.1.1 Educating People
			11.2.1.2 Organizational Transformation
			11.2.1.3 Decentralization of Work Networks
			11.2.1.4 Sustainability
		11.2.2 Integration of 4IR Technology
			11.2.2.1 Platform Capabilities
			11.2.2.2 No-Code Development Capabilities Scenario
			11.2.2.3 Low-Code Development Capabilities Scenario
			11.2.2.4 API-Based Ecosystem
	11.3 The Right Data Platform
		11.3.1 Machine Learning into 4IR
		11.3.2 Big Data and IoT
	11.4 Benefits of Smart City 4. 0
		11.4.1 Data-Based and Transportation
		11.4.2 Safer Communities in the Future
		11.4.3 Reduction of the Environmental Footprint and Increase in Digital Equity
		11.4.4 Newer and Greater Economic Opportunities
		11.4.5 Infrastructure Improvement
	11.5 Limitations of Smart City 4.0
	11.6 Conclusion and Future Aspects
	References
Chapter 12: Deep Learning Approach in Malware Hunting
	12.1 Introduction
		12.1.1 Deep Learning
		12.1.2 Attacks
	12.2 Learning Techniques
		12.2.1 Artificial Neural Network
		12.2.2 Convolutional Neural Network
		12.2.3 Recurrent Neural Network
		12.2.4 Multilayer Perceptron Neural Network
		12.2.5 Generative Adversarial Network
		12.2.6 Restricted Boltzmann Machine
		12.2.7 Deep Belief Network
	12.3 Activation Functions
		12.3.1 Role of Activation Function
		12.3.2 Types of Activation Function
			12.3.2.1 Binary Step Function
			12.3.2.2 Linear Activation Function
			12.3.2.3 Non-Linear Activation Functions
			12.3.2.4 Sigmoid/Logistic
			12.3.2.5 ReLU (Rectified Linear Unit)
			12.3.2.6 Leaky ReLU
			12.3.2.7 Parameter ReLU
			12.3.2.8 Softmax
			12.3.2.9 Swish
	12.4 Malware Attacks
		12.4.1 Introduction to Malware
		12.4.2 Categories of Malware
		12.4.3 Stages
			12.4.3.1 Static Properties Analysis
			12.4.3.2 Manual Code Reversing
	12.5 Deep Learning over Malware Attacks
		12.5.1 Techniques
			12.5.1.1 Signature-Based Detection
			12.5.1.2 Heuristic-Based Detection
			12.5.1.3 Static Analysis
			12.5.1.4 Dynamic Analysis
			12.5.1.5 Hybrid Analysis
	12.6 Result
	12.7 Conclusion
	References
	Blog
Chapter 13: Prediction of Breast and Lung Cancer, Comparative Review and Analysis Using Machine Learning Techniques
	13.1 Introduction
	13.2 Literature Review
	13.3 Machine Learning Techniques
		13.3.1 Logistic Regression Classifier
		13.3.2 Support Vector Machine (SVM) Classifier
		13.3.3 Naive Bayes Classifier
		13.3.4 Decision Tree Classifier
		13.3.5 k-Nearest Neighbor Classifier
	13.4 Process Flow Diagram
	13.5 Dataset and Implementation
		13.5.1 Breast Cancer Dataset
			13.5.1.1 Breast Cancer Dataset Attributes
			13.5.1.2 Dataset Visualization of Breast Cancer
			13.5.1.3 Class Visualization of Breast Cancer
			13.5.1.4 Correlation
			13.5.1.5 Implementation Snapshots ( Figure 13.10)
			13.5.1.6 Performance Metrics
			13.5.1.7 Accuracy Metrics ( Table 13.2)
			13.5.1.8 Accuracy Table
		13.5.2 Lung Cancer
			13.5.2.1 Attribute Information
			13.5.2.2 Dataset Visualization of Lung Cancer
			13.5.2.3 Class Label Visualization
			13.5.2.4 Accuracy Table
			13.5.2.5 ROC Curve
	13.6 Comparison and Analysis
	13.7 Conclusion
	References
Index