دانلود کتاب Proceedings of International Conference on Computational Intelligence and Data Engineering: ICCIDE 2022

فهرست مطالب :

Preface
Contents
A Review of Deep Learning Methods in Automatic Facial Micro-expression Recognition
1 Introduction
2 Micro-expression Datasets
2.1 Spontaneous Micro-expression Corpus (SMIC)
2.2 Chinese Academy of Sciences Micro-expression (CASME) Dataset
2.3 Chinese Academy of Sciences Micro-expression (CASME II) Dataset
3 Micro-expression Recognition
3.1 Preprocessing
3.2 Feature Extraction
3.3 Classification
4 Performance Metrics
5 Conclusion
References
Mathematical Modeling of Diabetic Patient Model Using Intelligent Control Techniques
1 Introduction
2 Literature Survey
3 Materials and Methods
4 Conclusion
References
End-to-End Multi-dialect Malayalam Speech Recognition Using Deep-CNN, LSTM-RNN, and Machine Learning Approaches
1 Introduction
2 Related Work
3 Proposed Methodology and Design
3.1 The Proposed Methodology
3.2 Dataset
3.3 Feature Extraction
3.4 Building the Accented ASR System
4 Experimental Results
5 Conclusion and Future Scope
References
JSON Document Clustering Based on Structural Similarity and Semantic Fusion
1 Introduction
2 Literature Survey
3 JSim
3.1 Schema Extraction
3.2 Similarity Computation
3.3 Clustering
4 Experimental Evaluation
4.1 Results
4.2 Discussion
5 Conclusions
References
Solar Power Forecasting to Solve the Duck Curve Problem
1 Introduction
2 Literature Review
3 Methodology
3.1 Duck Curve
3.2 Data Acquisition and Inputs
3.3 Flowchart
3.4 Machine Learning Algorithms Tested
4 Results
4.1 Performance of Machine Learning Models
4.2 Random Forest Performance Metrics
4.3 Calculation of Generated Power for VIT Chennai
4.4 Duck Curve for VIT Chennai
4.5 Case Study
5 Conclusion
References
Dynamic Optimized Multi-metric Data Transmission over ITS
1 Introduction
2 Related Multi-cast Mobility Models
3 Proposed Implementation
3.1 Procedure for selection of Multi-cast Node
3.2 Acknowledgment (ACK) Procedure in NOSMR
3.3 Representation of Acknowledgement Data Frame
4 Experimental Results
5 Conclusion and Future Work
6 Proof of the Concept
References
Solar Energy-Based Intelligent Animal Reciprocating Device for Crop Protection Using Deep Learning Techniques
1 Introduction
2 Real-World Problems of Image Classification
3 Test System Description
4 Methodology
4.1 Convolutional Neural Networks (CNNs)
4.2 Recurrent Neural Networks (RNNs)
5 Experimental Processing
6 Conclusion
References
Toward More Robust Classifier: Negative Log-Likelihood Aware Curriculum Learning
1 Introduction
2 Literature Survey
3 Uncertainty Estimation
3.1 Mathematical Formulas for Uncertainty Quantification
4 Negative Log-Likelihood and Uncertainty
4.1 Likelihood Versus Probability
4.2 Maximum Likelihood Estimation
4.3 Negative Log-Likelihood and Its Relationship with SoftMax Activations
5 Methodology
6 Results
7 Conclusion
References
Design of Fuzzy Logic Controller-Based DPFC Device for Solar-Wind Hybrid System
1 Introduction
2 Proposed Dynamic System
2.1 Concept of Wind Energy Conversion System
2.2 Concept of Solar PV System
3 Distributed Power Flow Controller
3.1 Central Control
3.2 Series Control
3.3 Shunt Control
4 DPFC with Fuzzy Logic Controller
5 Results and Discussion
5.1 Simulation Results of PV/Wind Hybrid System Without Any Custom Device
5.2 Simulation Results of PV/Wind Hybrid System with UPFC
5.3 Simulation Results of PV/Wind Hybrid System with DPFC
5.4 Simulation Results DPFC Device with Fuzzy Logic Controller
5.5 Comparative Analysis
6 Conclusion
References
Analysis of EEG Signal with Feature and Feature Extraction Techniques for Emotion Recognition Using Deep Learning Techniques
1 Introduction
2 Related Works
3 Methodology
3.1 Datasets
3.2 Preprocessing
3.3 Extraction of Features
3.4 Feature Selection (FS)
3.5 Classifier
4 Discussion and Results
5 Conclusion
References
Innovative Generation of Transcripts and Validation Using Public Blockchain: Ethereum
1 Introduction
2 Related Work
3 Proposed Methodology
3.1 Web Server Module
3.2 Certificate Verification Module
3.3 Raft Consensus Module
3.4 Inter-planetary File System
3.5 Hyperledger Fabric Channel Module
3.6 Hyperledger Fabric Transaction
3.7 Paxos Module
3.8 Organization Inclusion and Deletion
3.9 Feedback Module
3.10 Cloud Deployment Module
4 Experimental Results
4.1 Performance Metrics
5 Conclusion and Future Work
References
Windows Malware Hunting with InceptionResNetv2 Assisted Malware Visualization Approach
1 Introduction
2 Related Works
3 Methodology
3.1 Overview
3.2 Grayscale Malware Images
3.3 InceptionResNetv2 CNN
4 Experiments
4.1 Dataset Details
4.2 Evaluation Metrics
5 Results and Analysis
6 Conclusions and Future Scope
References
Custom-Built Deep Convolutional Neural Network for Breathing Sound Classification to Detect Respiratory Diseases
1 Introduction
1.1 Significance of the Proposed Model
1.2 Research Gap and Contributions of the Paper
2 Literature Survey
3 Methodology
4 Results and Discussions
5 Conclusions and Future Scope
References
Infrastructure Resiliency in Cloud Computing
1 Introduction
2 Methodology
2.1 System Set-Up Stage
3 Experimental Results
4 Conclusion
4.1 Limitations
4.2 Recommendation
4.3 Future Research
References
Deep Learning Model With Game Theory-Based Gradient Explanations for Retinal Images
1 Introduction
2 Materials and Method
2.1 Integrated Gradients
2.2 Smooth-Grad
2.3 Shapley Values
2.4 Proposed SHAP Gradient DR Model
3 Results and Discussion
3.1 Dataset Description
3.2 Performance Evaluation of DR Classification
3.3 Analysis of the Explainability SHAP Gradient DR Model
4 Conclusion
References
A Comparative Analysis of Transformer-Based Models for Document Visual Question Answering
1 Introduction
2 Related Work
2.1 Text-Based Visual Question Answering
2.2 Transformer-Based Natural Language Processing Models
2.3 Multimodal Learning in Vision and Language Tasks
3 Methodology
3.1 Method for Text Extraction
3.2 Method for Answer Extraction Transformer-Based Models
4 Experiments and Results
4.1 DocVQA
4.2 Experimental Setup
4.3 Results Obtained by Various Models on DocVQA
4.4 Experiments
5 Conclusion and Future Work
References
Develop Hybrid Wolf Optimization with Faster RCNN to Enhance Plant Disease Detection Performance Analysis
1 Introduction
2 Related Works
3 Proposed Method
3.1 Data Collection
3.2 Feature Extraction
3.3 Predicting Diseased Spots
4 Experimental Evaluation
4.1 Prediction of Accuracy and Speed
5 Conclusion
References
An Efficient CatBoost Classifier Approach to Detect Intrusions in MQTT Protocol for Internet of Things
1 Introduction
2 Related Work
3 Message Queuing Telemetry Transport (MQTT)
4 Proposed Method for CatBoost Classifier
4.1 Research Methodology
5 Results and Discussion
5.1 Imbalanced Dataset
6 Conclusion
References
Self-regulatory Fault Forbearing and Recuperation Scheduling Model in Uncertain Cloud Context
1 Introduction
1.1 Cloud Computing
1.2 Task Scheduling
1.3 Fault Tolerance
1.4 Types of Failures in Cloud Environment
2 Literature Review
2.1 Cloud and Its Uncertainties
2.2 Resource Provisioning Strategies for Improving QoS
2.3 Fault Tolerance and Detection Techniques
2.4 Fault Tolerance and Management Techniques
2.5 Challenges Identified
3 Proposed Methodology
3.1 Experimental Setup and Dataset Description
3.2 RFDR Model
3.3 Fault Injection Mechanism
3.4 Fault Detection Mechanism
3.5 Recovery Mechanism
3.6 Failure Metrics and QoS Parameters Used
4 Experimental Evaluation and Results
4.1 Analyzing Performance of RFDR in Various Scheduling Algorithms
5 Conclusion
References
A Comprehensive Survey on Student Perceptions of Online Threat from Cyberbullying in Kosova
1 Introduction
1.1 Cyberbullying Consequences
2 Related Work
3 Methodology
3.1 Research Questions
3.2 Limitations
3.3 Data Analysis Procedures
3.4 Data Presentation
3.5 Types of Questionnaire Questions
4 Results
5 Discussion
6 Future Work
7 Conclusions
References
Addressing Localization and Hole Identification Problem in Wireless Sensor Networks
1 Introduction
2 Technological Background
3 Proposed System
3.1 Addressing Localization
3.2 Hole Identification and Healing
3.3 Implementation and Result Analysis
4 Conclusion and Future Work
References
The Impact of ICMP Attacks in Software-Defined Network Environments
1 Introduction
2 Literature Review
3 Internet Control Message Protocol (ICMP)
3.1 Operations of ICMP Protocol
3.2 ICMP Messages
3.3 ICMP Message Types
3.4 ICMP Message Format
4 ICMP Attacks in SDN Environments
4.1 Application Layer Flooding
4.2 SYN Flooding
5 Experiment and Results
5.1 CPU Utilization
5.2 Control Channel Bandwidth
5.3 Packet Delivery Ratio
5.4 Flow Request Analysis
6 Conclusion
References
An Area-Efficient Unique 4:1 Multiplexer Using Nano-electronic-Based Architecture
1 Introduction
1.1 QCA Cell
1.2 QCA Majority Gate
1.3 QCA Inverter
2 Five Input Majority Gate
3 2:1 Multiplexer
4 Proposed 4:1 Multiplexer
5 Simulation and Experimental Result Analysis
5.1 Implementation of 2:1 and Proposed 4:1 Multiplexer
5.2 Complexity of Proposed 4:1 Multiplexer
5.3 Analysis of Proposed 4:1 Multiplexer
6 Conclusion
References
Digital Realization of AdEx Neuron Model with Two-Fold Lookup Table
1 Introduction
2 AdEx Model
3 Proposed AdEx Model Using Two-Fold Lookup Table
3.1 f-LUT and e-LUT Architecture
3.2 Bit Width Requirements and Compressibility
3.3 Data Retrieval from the Two-Fold Lookup Table
4 Hardware Implementation
5 Simulation Results
6 Conclusion
References
A Novel Quantum Identity Authentication Protocol Based on Random Bell Pair Using Pre-shared Key
1 Introduction
2 Basic Ideas on EPR and Entanglement Swapping
3 Proposed Protocol
3.1 General Procedure
3.2 Quantum Procedure
3.3 Classical Procedure
4 Security of Proposed Protocols
4.1 Eve Guess on Correct Bell States Position
4.2 Eve Guess on Wrong Bell States Position
5 Conclusion
References
Analysis of Hate Tweets Using CBOW-based Optimization Word Embedding Methods Using Deep Neural Networks
1 Introduction
2 Background Study
3 System Model
3.1 Word Embedding Technique
3.2 Problem Statement
3.3 Proposed Architecture
3.4 Optimised CBOW Techniques
4 Performance Evaluation
4.1 Pre-processing
4.2 Pre-processing Steps
4.3 Techniques to Deal with Unbalanced Data
4.4 Training and Testing Data
4.5 Evaluation Metrics
4.6 Comparison with Existing Methods
5 Conclusion
References
Performance Analysis of Discrete Wavelets in Hyper Spectral Image Classification: A Deep Learning Approach
1 Introduction
2 Related Work
3 Methodology
3.1 Wavelet CNN Framework
3.2 Feature Decomposition Using Discrete Wavelets
4 Experiments and Result Analysis
4.1 Classification Results
5 Conclusion
References
Tyro: A Mobile Inventory Pod for e-Commerce Services
1 Introduction
2 Related Work
3 Methodology
3.1 Mecanum Wheels
3.2 Husky Lens Camera Module
3.3 Proposed System
3.4 Working
4 Results and Discussion
5 Conclusion
References
Segmentation and Classification of Multiple Sclerosis Using Deep Learning Networks: A Review
1 Introduction
1.1 Stages and Types of MS
2 MRI, Datasets, and Evaluation Measures
2.1 MRI
2.2 Dataset
2.3 Evaluation Measures
3 Deep Learning (DL) Networks
3.1 DL Methods for MS Lesion Segmentation
3.2 CNN
3.3 U-Net
4 Lesion Classification
5 Issues and Challenges
6 Discussion
7 Conclusion
References
Malware Detection and Classification Using Ensemble of BiLSTMs with Huffman Feature Optimization
1 Introduction
2 Related Works
3 Methodology
3.1 Overview
3.2 Feature Extraction
3.3 Huffman Feature Optimization
3.4 Training Networks
4 Experiments
4.1 Dataset Details
4.2 Evaluation Metrics
5 Results and Analysis
6 Discussion
7 Conclusions and Future Scope
References
Detection of Location of Audio-Stegware in LSB Audio Steganography
1 Introduction
2 Related Work
3 Proposed System
3.1 LSB Location Clustering
3.2 ASCII Conversion
3.3 Location Finder
4 Experimental Setup
4.1 Dataset
4.2 Simulation Setup
4.3 Evaluation Metrics
5 Experimental Results and Discussions
5.1 Effectiveness of Proposed Algorithm in Identifying Audio-Stegware Location
6 Conclusion
References
Hybrid Quantum Classical Neural Network-Based Classification of Prenatal Ventricular Septal Defect from Ultrasound Images
1 Introduction
2 Methodology
2.1 Classical Convolution Neural Network Phase
2.2 Parameterized Quantum Circuit Phase
3 Results
4 Conclusion
References
Experimental Evaluation of Reinforcement Learning Algorithms
1 Introduction
2 Related Work
3 Methodology
3.1 Data Collection
3.2 Model Evaluation Approach
3.3 Training
4 Experiments
4.1 Computation Environment
4.2 Experiments Setup
4.3 Results
5 Conclusion
6 Future Work
References
An Approach to Estimate Body Mass Index Using Facial Features
1 Introduction
2 Literature Survey
3 Methodology
4 Result and Discussion
5 Conclusion
References
An Approach to Count Palm Tress Using UAV Images
1 Introduction
2 Literature Survey
3 Proposed Method
3.1 Methodology
3.2 Flowchart
4 Architecture
5 Results
6 Conclusions and Future Work
References
Comparative Analysis on Deep Learning Algorithms for Detecting Retinal Diseases Using OCT Images
1 Introduction
2 Literature Survey
3 Methodology
3.1 Role of Deep Learning
3.2 Segmentation Approaches in Detecting Retinal Diseases
3.3 Data Preprocessing Techniques in Detecting Retinal Images
4 Performance Metrics
5 Comparative Analysis
6 Conclusion
References
PCB-LGBM: A Hybrid Feature Selection by Pearson Correlation and Boruta-LGBM for Intrusion Detection Systems
1 Introduction
2 Related Works
3 Methodology
3.1 Dataset
3.2 Preprocessing
3.3 Hybrid Feature Selection
4 Results and Analysis
5 Comparison with Existing Methods
6 Conclusion
References
Extractive Summarization Approaches for Biomedical Literature: A Comparative Analysis
1 Introduction
2 Automatic Text Summarization
3 Related Works
4 Text Summarization Algorithms
4.1 Frequency-Based Approach
4.2 Similarity-Based Approach
4.3 Luhn Approach
5 Comparative Analysis of Text Summarization Approaches
6 Conclusion
References
SMS Spam Detection Using Federated Learning
1 Introduction
1.1 Objectives
2 Background and Literature Survey
3 Federated Learning
3.1 Federated Learning Overview
3.2 Federated Learning Reference Architecture
4 Proposed System Architecture-Federative Learning
4.1 Application Development for Messaging
4.2 Backend
4.3 Working Methodology
4.4 System Details
4.5 Messaging Application
4.6 Database
5 Results
6 Conclusion and Future Scope
References
Data Extraction and Visualization of Form-Like Documents
1 Introduction
2 Previous Work
3 Methodology
3.1 Image Preprocessing and Segmentation
3.2 Text Extraction
3.3 Sentiment Analysis
4 Results and Discussion
5 Future Work
6 Conclusion
References
Author Index