Advanced Soft Computing Techniques in Data Science, IoT and Cloud Computing

✍ Scribed by Sujata Dash (editor), Subhendu Kumar Pani (editor), Ajith Abraham (editor), Yulan Liang (editor)

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 443
• Series: Studies in Big Data, 89
• Edition: 1st ed. 2021
• Category: Library

✦ Synopsis

This book plays a significant role in improvising human life to a great extent. The new applications of soft computing can be regarded as an emerging field in computer science, automatic control engineering, medicine, biology application, natural environmental engineering, and pattern recognition. Now, the exemplar model for soft computing is human brain. The use of various techniques of soft computing is nowadays successfully implemented in many domestic, commercial, and industrial applications due to the low-cost and very high-performance digital processors and also the decline price of the memory chips. This is the main reason behind the wider expansion of soft computing techniques and its application areas. These computing methods also play a significant role in the design and optimization in diverse engineering disciplines. With the influence and the development of the Internet of things (IoT) concept, the need for using soft computing techniques has become more significant than ever. In general, soft computing methods are closely similar to biological processes than traditional techniques, which are mostly based on formal logical systems, such as sentential logic and predicate logic, or rely heavily on computer-aided numerical analysis. Soft computing techniques are anticipated to complement each other. The aim of these techniques is to accept imprecision, uncertainties, and approximations to get a rapid solution.

However, recent advancements in representation soft computing algorithms (fuzzy logic,evolutionary computation, machine learning, and probabilistic reasoning) generate a more intelligent and robust system providing a human interpretable, low-cost, approximate solution. Soft computing-based algorithms have demonstrated great performance to a variety of areas including multimedia retrieval, fault tolerance, system modelling, network architecture, Web semantics, big data analytics, time series, biomedical and health informatics, etc. Soft computing approaches such as genetic programming (GP), support vector machine–firefly algorithm (SVM-FFA), artificial neural network (ANN), and support vector machine–wavelet (SVM–Wavelet) have emerged as powerful computational models. These have also shown significant success in dealing with massive data analysis for large number of applications.

All the researchers and practitioners will be highly benefited those who are working in field of computer engineering, medicine, biology application, signal processing, and mechanical engineering. This book is a good collection of state-of-the-art approaches for soft computing-based applications to various engineering fields. It is very beneficial for the new researchers and practitioners working in the field to quickly know the best performing methods. They would be able to compare different approaches and can carry forward their research in the most important area of research which has direct impact on betterment of the human life and health.

This book is very useful because there is no book in the market which provides a good collection of state-of-the-art methods of soft computing-based models for multimedia retrieval, fault tolerance, system modelling, network architecture, Web semantics, big data analytics, time series, and biomedical and health informatics.

✦ Table of Contents

Preface
Acknowledgements
Contents
Editors and Contributors
Abbreviations
Soft Computing Techniques for IoT Devices
A Wearable Assistive Device for Safe Travel Using Transfer Learning and IoT for Visually Impaired People
1 Introduction
2 Review of Literature
3 Overview of Proposed System
4 Proposed Methodology
4.1 Bus Board Detection for Route Information Extraction
4.2 Route Recognition
4.3 Text to Speech Conversion for Route Identification
4.4 An IOT Based Railway Platform Detection
4.5 Assistive Device for Visual Mobility and Safe Travel
5 Implementation and Results
6 Conclusion and Future Work
References
Soft Computing Techniques for Physical Layer Security of IoT Devices
1 Introduction
2 Machine Learning Algorithms for PLS in IoT
2.1 PLS Schemes Using K-Nearest Neighbor Algorithm
2.2 PLS Scheme Using Support Vector Machine
3 Deep Learning Algorithms for PLS in IoT
3.1 PLS Scheme Using Deep Neural Network Models
3.2 PLS Scheme Using Convolutional Neural Network Models
3.3 PLS Scheme Using Long Short-Term Memory Networks
3.4 PLS Schemes Based on Recurrent Neural Networks
4 Fuzzy Systems for Physical Layer Security for IoT
5 Genetic Algorithm for Physical Layer Security for IoT
6 Conclusions
References
Linear Congruence Generator and Chaos Based Encryption Key Generation for Medical Data Security in IoT Based Health Care System
1 Introduction
2 Literature Survey
3 Problem Domain
4 Our Contributions
5 Proposed Work
5.1 Session Key Generation
5.2 Intermediate Key Generation
5.3 Encryption Process
5.4 Generation of Authentication Code and Transmission File
5.5 Decryption Phase
6 Result and Discussion
6.1 Randomness Analysis of Key
6.2 Comparative Analysis Between Modified Logistic Map and Standard Logistic Map
6.3 Statistical Analysis
6.4 Key Sensitivity Analysis
6.5 Security Analysis
6.6 Functionality Analysis of Encryption Technique
6.7 Significance of Authentication in Our Proposed Scheme
7 Conclusion
References
Content Based Video Retrieval—Methods, Techniques and Applications
1 Introduction
2 State-of-the-Art Techniques
3 Content Based Video Retrieval (CBVR)
3.1 Keyframe Extraction
3.2 Feature Vectors for CBVR
3.3 The Proposed Method
3.4 Performance Analysis
4 Applications
5 Conclusion
References
Building the World of Internet of Things
1 Introduction: The Information Revolution
2 Sensors
3 Networks
4 Augmented Intelligence
5 Augmented Behavior
6 Standards
7 Conclusion
References
Applicability of Machine Learning Algorithms for Intelligent Farming
1 Introduction
2 Literature Review
3 Data Analysis and Its Techniques Used in Machine Learning
3.1 Feature Selection
4 Background Statistics of ML Algorithms
4.1 Chi-Square Statistics
4.2 Euclidean Distance
5 Classification Algorithms Used
5.1 K-Nearest Neighbor’s
5.2 Support-Vector Machine (SVM)
5.3 Logistic Regression
5.4 Decision Trees
6 Implementation
6.1 Gathering Data
6.2 Analyzing Gathered Data
6.3 Results
6.4 Integrating IoT and ML for Intelligent Farming—A Future Approach
7 Conclusion and Future Scope
References
Soft Computing Techniques in Cloud Computing and Computer Networking
Hybrid Cloud Data Protection Using Machine Learning Approach
1 Introduction
2 Problem Statement
3 De-duplication
4 Cloud Security Protection Framework with Machine Learning Modules
4.1 Cloud Client Classification Using Enhanced C4.5 Algorithm
4.2 De-duplication Processing Algorithm
4.3 Dynamic Spatial RBAC Algorithm
5 Experimental Results
6 Summary
References
Analysis of Long Short Term Memory (LSTM) Networks in the Stateful and Stateless Mode for COVID-19 Impact Prediction
1 Introduction
2 Recurrent Neural Networks
2.1 From RNN to LSTM
3 LSTM Architecture
3.1 Various Gates in the LSTM Architecture
3.2 Stateful and Stateless LSTM
4 LSTM Research
5 COVID19-Prediction Problem
6 Defining Models in Keras
6.1 Five Step Lifecycle
7 LSTM State Management
8 Results and Conclusion
8.1 A Vanilla RNN
8.2 Stateful LSTM
8.3 Stateless LSTM Without Shuffling
8.4 Stateless with Shuffling
References
Soft Computing Techniques for Energy Consumption and Resource Aware Allocation on Cloud: A Progress and Systematic Review
1 Introduction
2 Motivation
3 Background
3.1 Framework for Energy Consumption and Resource Aware Allocation
3.2 Types of Soft Computing Techniques
3.3 Importance of Soft Computing Techniques for Energy Consumption
3.4 Research Challenges or Issues
3.5 Application Areas
4 Reported Work
4.1 Soft Computing Techniques for Energy Consumption
4.2 Resource Allocation
5 Comparative Analysis of Soft Computing Techniques for Energy Consumption and Resource Aware Allocations
6 Conclusion
References
Automatic Segmentation and Classification of Brain Tumor from MR Images Using DWT-RBFNN
1 Introduction
2 Literature Review
3 Materials and Methods
3.1 Magnetic Resonance Imaging (MRI)
3.2 MRI Database
3.3 Wavelet Transform (WT)
3.4 DWT Image Filtering
4 Preprocessing of MRI Data
4.1 Resizing of MRI Images
4.2 Image Denoising Using MRDWT-ICA
4.3 Tumor Segmentation
4.4 Otsu’s Thresholding
5 RBFNN Classifier
5.1 Performance Evaluation Assessment Metrics (PEAM)
6 Experimental Result and Discussion
6.1 BMRI Image Segmentation Process
6.2 Feature Extraction
6.3 Feature Normalization
6.4 Data Distribution
6.5 Classification Result
6.6 Result Analysis
7 Conclusion and Future Scope
References
Automatic Localization of Optic Disc in Retinal Fundus Image Based on Unsupervised Learning
1 Introduction
2 Related Work
3 Clustering Methods
3.1 Agglomerative Clustering
3.2 K-Means Clustering
3.3 Fuzzy C Mean Clustering
3.4 DBSCAN Clustering
4 Methodology
4.1 Retinal Image Pre-processing
5 Results and Discussion
5.1 Dataset Description
5.2 Result Analysis
5.3 Optic Disc Detection
5.4 Performance Evaluation
5.5 Discussion
6 Conclusion
References
Soft Computing Techniques in Data Science
Performance Evaluation of Hybrid Machine Learning Algorithms for Medical Image Classification
1 Introduction
2 Methodologies
2.1 Efficiency of SVM-RBF Kernels for Medical Image Classification
2.2 Feature Reduction using PCA
2.3 Hybrid Algorithm Based on PSO, GA with Local Search
2.4 Deep Learning
3 Discussion of Experimental Results
4 Conclusion
References
Computing Truth Values of Modus Ponens and Modus Tollens Rule for Linguistic Truth-Valued Propositions and Its Application in Taking Decisions in Health Care
1 Introduction
2 Basic Concepts
3 Reasoning with Linguistic Truth-Valued Propositions
3.1 Computation of Truth Values of Modus Ponens and Modus Tollens rules for LTVP
3.2 Computation of Truth Values of Modus Ponens and Modus Tollens rules for QLTVP
4 Output and Results
4.1 Examples
5 Conclusion
References
Analysis of Customers’ Reviews Using Soft Computing Classification Algorithms: A Case Study of Amazon
1 Introduction
2 Literature Survey
3 Data Collection and Methodology
3.1 Data Collection and Preprocessing
3.2 Data Representation
3.3 Classifications
4 Experimental Results and Discussions
5 Conclusions
References
Pattern Mining—FTISPAM Using Hybrid Genetic Algorithm
1 Introduction
2 Fuzzy Time-Interval Sequential Patterns Using Hybrid Genetic Algorithm
3 Overview of the FTISPAM-HGA
4 Fuzzy Time Interval Sequential Pattern Mining
5 Hybrid Genetic Algorithm
6 Fuzzy Time Interval Sequential Pattern Mining Using HGA Algorithm
7 Patterns Matching Using SCI
8 Significant Pattern Evaluation
9 Experimental Results and Discussion
10 Conclusion
References
Soft Computing Techniques for Medical Diagnosis, Prognosis and Treatment
1 Introduction
1.1 Healthcare Data
1.2 Types of AI in Healthcare
2 Intelligent Systems for Healthcare Decisions
2.1 Virtual Assistants in Drug Development
2.2 Intelligent Medical Devices
2.3 Ambient Healthcare Monitoring System
3 Use Cases of Soft Computing in Basic Sciences and Diagnostics
3.1 Soft Computing in Basic Sciences
3.2 Soft Computing in Medical Diagnosis
4 Soft Computing Techniques in Healthcare Decision Systems
4.1 Artificial Neural Networks
4.2 Fuzzy Logic
4.3 Genetic Algorithms
5 Further Applications of Soft Computing in Healthcare Decision Making
5.1 Fuzzy Logic in Remote Healthcare Monitoring
5.2 Risk Assessment of Cervical Cancer in Women-Based on Convolutional Neural Network
5.3 Diagnosis of Depression Using Neuro-fuzzy Model of Soft Computing
6 Hybrid Techniques Used in Healthcare
6.1 Hybrid Solution for Skin Cancer Detection
7 Soft Computing in Clinical Applications
7.1 Soft Computing in Cardiology
7.2 Soft Computing in Neurology
7.3 Soft Computing in Medicine and Rehabilitation
7.4 Soft Computing in Other Clinical Areas
8 Conclusion
References
Role of Artificial Intelligence in COVID-19 Pandemic
1 Introduction
2 Premature Detection of the Coronavirus (COVID-19)
3 Succinct Review on Transferable Syndrome Outburst in the Year 2020
4 Applications of Artificial Intelligence in COVID-19 Pandemic
4.1 Premature Detection and Diagnosis of Infection
4.2 Protrusion of Suitcases and Transience
4.3 Progress of Drugs and Vaccines
4.4 Tumbling the Work of Healthcare Employees
5 The Original AI Capability of Bluedot and Metabiota
5.1 Bluedot
5.2 Metabiota Metadata
6 Conclusion
References
Prediction of Transmittable Diseases Rate in a Location Using ARIMA
1 Introduction
2 Related Work
3 Methodology
4 Comparative Study
5 Results and Discussion
6 Conclusion
References