Artificial Intelligence for Cloud and Edge Computing

Preface
Contents
Editors' Biography
An Optimization View to the Design of Edge Computing Infrastructures for IoT Applications
1 Introduction
2 IoT Applications: Classification and Challenges
2.1 A Classification by Field of Application
2.1.1 Automotive
2.1.2 Industry 4.0
2.1.3 E-Health
2.1.4 Smart Cities Applications
2.1.5 Retail
2.1.6 Smart Agriculture
2.2 Challenges of IoT Applications
3 Literature Review
4 Performance Modeling in Edge Computing Infrastructures
4.1 Efficiency Metrics
4.2 Performance Metrics
5 Problem Formulation
5.1 Model Parameters
5.2 Objective Functions and SLA
5.3 Optimization Problem
6 Heuristics
6.1 Variable Neighborhood Search
6.2 Genetic Algorithm
6.2.1 Problem Definition
6.2.2 Chromosome Encoding
6.2.3 Genetic Operators
7 Experimental Evaluation
7.1 Experimental Scenario
7.2 Experimental Results
8 Conclusion and Future Work
References
AIOps: A Multivocal Literature Review
1 Introduction
2 Background
2.1 Artificial Intelligence
2.2 Artificial Intelligence for IT Operations (AIOps)
3 Research Methodology
3.1 Research Questions
3.2 Search Strategy
4 Results
4.1 Trends
4.2 Definition of AIOps (RQ1)
4.3 Benefits of AIOps (RQ2)
4.4 Challenges of AIOps (RQ3)
4.5 Limitations and Potential Threat to Validity
5 Conclusions and Future Work
A.1 Appendix A: List of Sources Included in the MLR
References
Deep Learning-Based Facial Recognition on Hybrid Architecture for Financial Services
1 Introduction
2 Related Work
2.1 Facial Recognition
2.2 Edge–Cloud Computing
3 Methods
3.1 Facial Recognition Methods
3.2 Hybrid Architecture
4 Experimental Results
5 Conclusion
References
Classification of Swine Disease Using K-Nearest Neighbor Algorithm on Cloud-Based Framework
1 Introduction
2 Background and Literature Review
2.1 Related Works
3 Materials and Methods
3.1 Collection of Datasets
3.2 Proposed Algorithm
3.3 Cloud Platform
4 Results and Discussion
4.1 Home Page Interface
4.2 List of Diseases
4.3 Symptoms List
4.4 Treatment
5 Conclusion and Future Works
References
Privacy and Trust Models for Cloud-Based EHRs Using Multilevel Cryptography and Artificial Intelligence
1 Introduction
2 Related Works
2.1 Review of Selected Related Research Works
3 The Security/Privacy Mechanism
3.1 The Privacy Model
3.1.1 The Advanced Encryption Standard (AES)
3.1.2 The Elliptic Curve Cryptography (ECC)
3.2 The Access Control Aspect of the Security Mechanism
4 The Trust Architecture
4.1 Trust Concept
4.2 Managing Trust
4.3 The Trust Management Procedure (TMP)
5 The Proposed System Design Methodology
6 Results and Discussions
6.1 Performance Evaluation of the Proposed EHR System
6.2 Comparison of the Proposed System and Existing Works as per Trust
7 Conclusion and Future Work
References
Utilizing an Agent-Based Auction Protocol for Resource Allocation and Load Balancing in Grid Computing
1 Introduction
2 Distributed Artificial Intelligence
2.1 Benefits of Artificial Intelligence Load Balancing in Grid Computing
3 Related Works
4 Using Auction Model for Design: A Load Balancing System Based on Artificial Intelligence in Grid Computing
5 Artificial Intelligent Agents Interactions
6 Experimental Environment
6.1 System Implementation
6.2 Simulated Parameters
7 Performance Evaluation
8 Conclusion and Future Works
References
Optimization Model of Smartphone and Smart Watch Based on Multi Level of Elitism (OMSPW-MLE)
1 Introduction
2 Complex Network
2.1 Characterizing Networks
3 Design OMSPW-MLE
3.1 Main Stage of OMSPW-MLE
3.1.1 Data Collection and Preprocessing Stage
3.1.2 Apply Deterministic Selection Algorithm (Dselect)
3.1.3 Apply Ant Lion Optimization Algorithm (ALO)
4 Results and Analysis
4.1 Implementation OMSPW-MLI Stages
4.1.1 Description of Database
4.1.2 Draw Network
4.1.3 The Results of Dselect Algorithm
4.1.4 Results of ALO Algorithm
4.1.5 Evaluation
5 Discussion
6 Conclusions and Recommendation of Future Works
References
K-Nearest Neighbour Algorithm for Classification of IoT-Based Edge Computing Device
1 Introduction
2 Background and Literature Review
2.1 Related Works
3 Materials and Methods
3.1 Datasets
3.2 K-Nearest Neighbor (KNN) Algorithm
4 Result and Discussion
4.1 System Evaluation
4.2 System Comparison
4.3 Discussion
5 Conclusion
References
Big Data Analytics of IoT-Based Cloud System Framework: Smart Healthcare Monitoring Systems
1 Introduction
2 Internet of Things in Healthcare System
3 The Edge Computing and Artificial Intelligence in Healthcare System
4 Internet of Things and Big Data Analytics
5 Related Work
6 The Architecture of Big Data Analytics of IoT-Based Cloud Healthcare Monitoring System
6.1 The Flowchart of the Proposed System
6.2 The Architecture of the Healthcare Monitoring System
6.3 The Data Generation Layer
6.4 Data Processing
6.5 Application and Monitoring Management Layer
7 Results and Discussions
8 Conclusion and Future Research Direction
References
Genetic Algorithm-Based Pseudo Random Number Generation for Cloud Security
1 Introduction
2 Background and Related Work
3 Proposed System
3.1 Hybrid Model with a Combination of a Genetic Algorithm and LFSR
3.1.1 Algorithm for Generating Pseudo-Random Numbers
3.1.2 Design of LFSR
3.1.3 Genetic Algorithm (GA)
3.2 CloudSim
3.3 Proposed Hybrid Model Design
3.4 Key Sequence Generated Using GA
3.5 Statistical Test Results of Hybrid Model PRNG
3.5.1 Maximum Length
3.5.2 Test for Uniformity
3.5.3 Test for Independency/Autocorrelation
3.5.4 Run Test
4 Implementatıon Detaıls
4.1 Employment of Hybrid Model for the Encryption of Image Data over the Cloud System
5 Result Analysis of the Cipher System Based on Security Parameters
5.1 Security Constraints
5.1.1 Computation of Standard Deviation on the Occurrence of Ciphertext Elements
5.1.2 Entropy Computation of Elements of Ciphertext
5.2 Result Analysis of the Cipher System Based on Security Constraints
6 Conclusions and Future Scope
References
Anomaly Detection in IoT Using Machine Learning
1 Introduction
2 Literature Review
3 Methodology
3.1 Proposed Model
3.2 Hardware Setup
3.3 Data Preparation
3.4 Labelling Data
4 Algorithm Implementation and Parameter Tuning
4.1 SVM Implementation
4.2 Decision Tree Implementation
4.3 K Nearest Neighbor Implementation
4.4 Random Forest Implementation
4.5 Logistic Regression Implementation
5 Result and Analyses
5.1 Accuracy and Relative Time to Run for Each Model
5.2 Confusion Matrices of Each Model
5.3 Area of ROC Curve for Each Model
6 Conclusion and Future Work
References
System Level Knowledge Representation for Edge Intelligence
1 Introduction
2 Background
3 AI as a Spectrum
4 Complexity in Edge Intelligence
4.1 Seeing the Fog for the Cloud
4.2 Networks of Things and People
4.3 Sensors
4.4 Interfaces
4.5 The Sensor Mesh
4.6 Selection Automation
4.7 Physical Transport Layer
4.8 Real-Time Data Streams
4.9 Personal Storage Devices and the Privacy Layer
5 The Role of KR in IoT
5.1 KR to Capture Complexity
5.2 The Knowledge System Level
5.3 System Level KR as an Object (KRO)
5.4 Model Validation
6 Conclusion
References
AI-Based Enhanced Time Cost-Effective Cloud Workflow Scheduling
1 Introduction
2 Problem Description
2.1 Summary of the Mathematical Model – From [10]
2.2 Summary of Definitions – From [10]
3 QoS-Based Pricing Plan
4 Scheduling Algorithms
4.1 Versatile Time-Cost Algorithm (VTCA)
4.2 AI-Based Time-Constrained Early-Deadline Cost-Effective Algorithm (TECA)
5 Experimental Analysis
6 Conclusions
References
AI-JasCon: An Artificial Intelligent Containerization System for Bayesian Fraud Determination in Complex Networks
1 Introduction
2 Literature Review
2.1 Existing Research Efforts
2.2 Limitations of Exiting Systems
3 Fraud Detection System Model
4 System Implementation
5 System Design and Implementation
6 Conclusion
References
Performance Improvement of Intrusion Detection System for Detecting Attacks on Internet of Things and Edge of Things
1 Introduction
2 Related Work
3 Materials and Methodology
3.1 Data Preprocessing
3.2 Normalization Technique
3.3 Feature Selection
3.4 Principal Component Analysis
3.5 K-Nearest-Neighbor
3.6 Decision Tree
3.7 Gradient Boosting Machine
3.8 Extreme Gradient Machine
3.9 Light Gradient Machine
4 Results and Discussion
4.1 Evaluation Dataset
4.2 Performance Metrics of the Models on Sixfold Cross-Validation
4.3 Performance Metrics of the Models on Tenfold Cross-Validation
4.4 Comparison of the Proposed Models Utilizing Sixfold Cross-Validation with the State-of-the-Art Methods
4.5 Comparison of the Proposed Models Utilizing Tenfold Cross-Validation with the State-of-the-Art Techniques
5 Conclusion and Future Work
References
Index