Deep Sciences for Computing and Communications: First International Conference, IconDeepCom 2022 Chennai, India, March 17–18, 2022 Revised Selected Papers

✍ Scribed by Kottilingam Kottursamy, Ali Kashif Bashir, Utku Kose, Annie Uthra

• Publisher: Springer
• Year: 2023
• Tongue: English
• Leaves: 375
• Series: Communications in Computer and Information Science, 1719
• Category: Library

✦ Synopsis

This book constitutes selected papers presented during the First International Conference on Deep Sciences for Computing and Communications, IconDeepCom 2022, held in Chennai, India, in March 2022.
The 27 papers presented were thoroughly reviewed and selected from 97 submissions. They are organized in topical sections as follows: ​classification and regression problems for communication paradigms; deep learning and vision computing; deep- recurrent neural network (RNN) for industrial informatics; extended AI for heterogeneous edge.

✦ Table of Contents

Preface
Organization
Contents
Classification and Regression Problems for Communication Paradigms
Comparative Analysis of Book Recommendation System Based on User Reviews Using Hybrid Methods
1 Introduction
2 Literature Review
3 Proposed System
3.1 Grouping of Reviewers
3.2 Recommendation of Books
4 Result and Analysis
4.1 Dataset Description
5 Conclusion
References
Health Care Support for Elderly People in Resource Poor Areas
1 Introduction
2 Related Work
3 Proposed Work
3.1 Classification Using Support Vector Machine (SVM)
3.2 Classification Using Random Forest Algorithm
3.3 Results and Discussion
4 Conclusion
References
Finding Best Voting Classifier for Diabetic Disease Classification
1 Introduction
2 Literature Survey
3 Methodology
3.1 Data Description
3.2 Data Pre-processing
4 Model Architecture
4.1 Ensemble Learning
4.2 Voting Classifiers
4.3 Logistic Regression
4.4 Naïve Bayes
4.5 Random Forest
4.6 Ada Boost
4.7 Decision Tree
4.8 Workflow
5 Experimental Results
6 Conclusion
References
Convolutional Neural Networks for Traffic Sign Classification Using Enhanced Colours
1 Introduction
1.1 Problem Description
2 Literature Review
3 Proposed System
3.1 Dataset Collection
3.2 Data-Preprocessing
3.3 Classification
3.4 Algorithms
3.5 Results and Discussion
4 Conclusion
References
A Systematic Review on Fog Computing Security Algorithms on Current IoT Applications and Solutions
1 Introduction
2 Major Fog Computing Platforms
2.1 Edge Computing
2.2 Cloudlet
2.3 Microdata Center
2.4 Network Level Virtualization and Fog Computing
3 Various Application Domains of Fog Computing and Its Security Issues
3.1 Web Optimization Using Fog Computing
3.2 Provisioning 5G Mobile Networks with Fog Computing
3.3 Amplifying Healthcare Systems Using Fog Computing
3.4 Video Surveillance with Fog Computing
3.5 Vehicular Ad-Hoc Network and Fog Computing
3.6 Cyber-Physical Systems and Fog Computing
3.7 Assortment and Pre-processing of Speech Data Using Fog Computing
4 Conclusion
References
Enhanced and Effective Ensemble Model for Breast Cancer Detection and Prediction in Modern Medical System in Machine Learning
1 Introduction
2 Literature Review
3 Proposed Work
3.1 Overview
3.2 Proposed System Architecture
3.3 Data Pre-processing
3.4 Principal Component Analysis
3.5 Classification and Prediction
4 Results and Discussions
4.1 Dataset Description
4.2 Kernel Distribution Plot
4.3 Elbow Plot
4.4 Proposed Ensemble Classifier
5 Conclusion
References
Deep Learning and Vision Computing
Hand Gesture Recognition for Human-Computer Interaction Using Computer Vision
1 Introduction
1.1 Hand Tracking Module
1.2 AI Virtual Mouse
1.3 Gesture-Based Volume Controller
1.4 Software Requirements
2 Literature Survey
3 System Architecture and Design
3.1 Architecture Block Diagram
3.2 Design of Modules
4 Methodology
4.1 Hand Tracking Module
4.2 AI Virtual Mouse
4.3 Gesture-Based Volume Controller
4.4 Utility Functions
5 Output
5.1 Hand Tracking Module
5.2 AI Virtual Mouse
5.3 Gesture-Based Volume Controller
6 Results and Discussions
7 Challenges
8 Conclusion and Future Enhancement
References
Deep Learning Taxonomy on Human Face Expression Recognition for Communication Applications
1 Introduction
2 Stages in FER
2.1 Preprocessing
2.2 Face Detection
2.3 Facial Component Detection
2.4 Feature Extraction
2.5 Classification
3 Strategic Survey on FER
4 Conclusion
References
An IoMT Based Prediction Model for Cardiac Monitoring Using Machine Learning Algorithms
1 Introduction
2 Literature Review
3 Proposed Methodology
3.1 Data Acquisition
3.2 Pre-processing
3.3 Feature Extraction and Selection
3.4 Classification using ML Algorithms
3.5 Prediction
4 Results and Discussion
4.1 Dataset Description
4.2 Pre-processing
4.3 Feature Extraction and Selection
5 Comparative Analysis
5.1 Classification Algorithms
6 Conclusion and Future Work
References
Circular BlockEntropies and Fast Legendre Moments Based Content Preserving Image Authentication
1 Introduction
2 Representation of Circular Blocks
3 Entropy
4 Fast Legendre Moments
5 Proposed Authentication System
5.1 Preprocessing
5.2 Local Features Based on Circular Block Entropies
5.3 Global Features Based on Fast Legendre moments
5.4 Hash Generation
5.5 Hash Verification
6 Experimental Results
6.1 Localization and Recovery
7 Conclusion
References
Predictive Model for Depression and Anxiety Using Machine Learning Algorithms
1 Introduction
1.1 Depression and Anxiety
1.2 Proposed Hybrid Model
2 Related Works
3 Proposed Methodology
3.1 WESAD Data Set
3.2 Data Collection
4 Results and Discussion
4.1 Logistic Regression
4.2 LDA Classification
4.3 QDA Classification
4.4 KNN
4.5 Decision Tree
5 Conclusion and Future Enhancement
References
A Deep Learning Based Species Reciprocal System for Partridge
1 Introduction
2 Literature Review
3 Related Work
4 Proposed System
4.1 Convolutional Layers
4.2 Pooling Layers
4.3 Implementation
4.4 Loading the Dataset
4.5 Model
4.6 Training
4.7 Saving the Model
4.8 Execution Procedure
5 Results
6 Performance Evaluation
7 Comparison
8 Conclusion and Future Work
8.1 Conclusion
8.2 Future Work
References
Detecting Pneumonia from the CT-Scan Images Using Convolutional Neural Networks and Transfer Learning Techniques
1 Introduction
2 Related Works
3 Dataset Description
3.1 Train Set
3.2 Test Set
3.3 Validation Set
4 Proposed Model
4.1 Standard CNN Model
4.2 ResNet
4.3 Inception Net
4.4 Batch Normalization Layer
4.5 Max Pooling
4.6 Dropout Layer
4.7 Output Layer
5 Proposed Approach and Architecture
5.1 Image Augmentation
5.2 Architecture
5.3 Sigmoid
6 Parameters, Training and Results
6.1 Parameters
6.2 Training
6.3 Results
7 Discussions
8 Conclussion
1. References
Emotion Recognition of People Based on Facial Expressions in Real-Time Event
1 Introduction
2 Related Work
3 Proposed Work
3.1 Multi-task Cascaded Convolutional Neural Network (MTCNN)
3.2 Convolutional Recurrent Neural Network (CRNN)
4 Structural Architecture
5 Output Snapshot
6 Conclusion
References
Deep-Recurrent Neural Network (RNN) for Industrial Informatics
A Novel Modified LSTM Deep Learning Model on Precipitation Analysis for South Indian States
1 Introduction
2 Related Work
3 Proposed Methodology
3.1 Dataset Collection and Preprocessing
3.2 Proposed Algorithm
4 Experimental Design and Analysis
4.1 Support Vector Machine
4.2 Random Forest
4.3 Naive Bayes
4.4 Genetic Algorithm
5 Conclusion and Future Work
References
Industrial Machine Learning Cloud Applications: Integrity and Confidentiality of Machine Sensitive Data
1 Introduction
2 Contribution
3 Prior Works
4 Preliminaries
4.1 Machine Learning Used Confidential (MLC) Protocol
5 Result and Discussion
6 Conclusion
References
Classification of Dermoscopy Textures with an Ensemble Feedback of Multilayer Perceptron
1 Introduction
2 Literature Review
3 Methodology
3.1 Multi-layer Perceptron
3.2 Activation Function
3.3 Ensemble Model
4 Proposed System
5 Results and Discussion
6 Conclusion
References
Monitoring and Detection of Plant Diseases Using Neural Networks
1 Introduction
2 Background
2.1 Image Processing
2.2 Converting Color to Grayscale
2.3 HSV Color Space for Feature Generation
2.4 Feature Extraction Using SURF Method
2.5 Classification by Neural Networks
3 Dataset
4 Proposed System
4.1 Module Description
5 Results and Discussions
5.1 Output of Affected Arachis Hypogaea (Groundnut Leaves)
6 Performance Analysis
7 Conclusion
References
A Review on Detection and Diagnosis of Melanoma Carcinoma Using Deep Learning
1 Introduction
2 Skin Cancer
2.1 Deep Learning (DL) Techniques
3 Dataset
4 Distribution of Classes
5 Literature Survey
6 Skin Lesion Detection in Melanoma Skin Cancer
7 Skin Lesion Segmentation in Melanoma Skin Cancer
8 Classification in Melanoma
9 Conclusion
References
Future of E-commerce by Implementing Blockchain Payments System
1 Introduction
2 Literature Survey
2.1 User Authentication
2.2 User Interface
2.3 Blockchain Technology
3 Proposed Method
3.1 Logistic Partner
3.2 Regulator
3.3 Transaction Handler
3.4 Seller
4 Significance
4.1 Transaction Per Second (TPS)
4.2 Database Security Enhancement
5 Architecture
5.1 Transaction
5.2 Signing
5.3 Data Encryption Key
5.4 Verification
5.5 Transaction Complete
6 Applications
6.1 Transaction Management
6.2 Data Security
7 Conclusion
References
Cluster-Centric Based Hybrid Approach for Cricket Sports Analytics Using Machine Learning
1 Introduction
2 Literature Review
3 Proposed Methodology
3.1 Batsmen Algorithm
3.2 Bowler Algorithm
4 Result
4.1 Dataset Description
4.2 Feature Selection
4.3 Evaluation measures
4.4 Deployment
5 Result and Analysis
6 Future work and Conclusion
References
Extended AI for Heterogeneous Edge
PPA Based MAC Unit Using Vedic Multiplier and XOR Logic
1 Introduction
2 Performance – Oriented Multipliers
2.1 Binary Multiplier
2.2 Wallace Multiplier
2.3 Booth Multiplier
3 Vedic Multiplication Scheme
3.1 Urdhva-Tiryakbhyam (UT) Sutra
3.2 Urdhva-Tiryagbhyam Sutra for Binary Numbers
4 Design of MAC Unit
4.1 Vedic Multiplier
4.2 MAC Unit
4.3 XOR Logic
5 Results and Discussion
5.1 DC Compiler PPA Results
5.2 Comparison of Power and Area
6 Conclusion
References
Nowcasting - Rain Intensity Prediction Using Machine Learning Models with Markov Chain Process
1 Introduction
2 Materials and Methods
2.1 Nowcasting Using Machine Learning Models
2.2 Nowcasting Using Machine Learning Models with Markov Chain Process
3 Results and Discussions
4 Conclusion
References
A Survey of WSN Efficient Network Restoration Model Using Genetic Algorithms
1 Introduction
2 Network Failure Types
2.1 Actor Failure
2.2 Connectivity
2.3 Internode Connectivity
2.4 Single Node Failure
3 Related Works
3.1 Proactive Methods
3.2 Reactive Methods
4 Experimental Analysis
5 Conclusion
References
An Evaluation on the Performance of Privacy Preserving Split Neural Networks Using EMNIST Dataset
1 Introduction
1.1 Image Classification Using Neural Networks
1.2 Distributed Collaborative Learning
1.3 Summary of Literature
2 Problem Formulation
2.1 Learning Model Selection
2.2 Popular Datasets
2.3 The EMNIST Dataset
2.4 Research Objectives
3 Methodology
3.1 Split Neural Networks Training
3.2 Federated Training
4 Experimental Results and Analysis
4.1 Model and Implementation of Federated Learning
4.2 Model and Implementation of Split Neural Networks
4.3 Result Analysis
5 Conclusion and Future Research
References
The Security Algorithm BR22-01 Used to Secure the COVID’19 Health Data
1 Introduction
2 Related Work
3 Methodology
4 Result and Discussion
5 Conclusion
References
Modelling Air Pollution and Traffic Congestion Problem Through Mobile Application
1 Introduction
2 State of the Art of Existing Work
3 Requirement Gathering
3.1 Prototyping
3.2 Interactions
3.3 Focus Groups
4 Requirement Analysis
4.1 Functional Requirements
5 Overall Framework for Proposed Model
5.1 Customer App
5.2 Merchant App
5.3 Scanner
5.4 Database Server
5.5 Fuel Limit Calculator
5.6 Fuel Calculator
5.7 CO2 Emission Prediction Model
5.8 Payment Portal
5.9 Bill Generator
6 Implementation Modules
6.1 MC CALl’s Quality Factors
6.2 Product Revision Factors
6.3 Product Transition Factors
7 Results and Discussions
8 Conclusion and Future Enhancement
References
Author Index