Advances in Distributed Computing and Machine Learning: Proceedings of ICADCML 2023

Description

This book is a collection of peer-reviewed best selected research papers presented at the Fourth International Conference on Advances in Distributed Computing and Machine Learning (ICADCML 2023), organized by Department of Computer Science and Engineering, National Institute of Technology, Rourkela, Odisha, India, during 15–16 January 2023. This book presents recent innovations in the field of scalable distributed systems in addition to cutting edge research in the field of Internet of Things (IoT) and blockchain in distributed environments.     Preface Contents About the Editors NS3-Based Performance Assessment of Routing Protocols AODV, OLSR and DSDV for VANETs 1 Introduction 2 Related Work 3 Routing Protocols 3.1 AODV 3.2 OLSR 3.3 DSDV 4 Performance Evaluation 4.1 Simulation Metrics 4.2 Simulation Setup 4.3 Result Analysis 5 Conclusion References A Novel Blockchain-Based Smart Contract for Real Estate Management 1 Introduction 2 Literature Survey 3 Blockchain-Based Smart Contract 4 Proposed Model 5 Discussion 6 Conclusion References A Review on VM Placement Scheme Using Optimization Algorithms 1 Introduction 2 Virtual Machine Placement 3 The Single-Objective Optimization Algorithms 3.1 Ant Colony Algorithm 3.2 Particle Swarm Optimization 3.3 Fireflies Algorithm 3.4 Honeybee Algorithm 3.5 Cuckoo Search 3.6 Genetics Algorithm 4 The Multi-objective Optimization Algorithms 4.1 Bat Algorithm 4.2 Bio-geography-Based Optimization VMP Schemes 5 Conclusion References Use of Blockchain to Prevent Distributed Denial-of-Service (DDoS) Attack: A Systematic Literature Review 1 Introduction 2 Research Questions and Article Selection 2.1 Research Questions 2.2 Inclusion Criteria 2.3 Inclusion Criteria 2.4 Manual Selection 2.5 Final Article Selection 3 Attribute Framework 3.1 Attribute Identification 3.2 Characterization of the Articles 4 Article Assessment and Review Results 5 Conclusion References CS-Based Energy-Efficient Service Allocation in Cloud 1 Introduction 2 Related Research Work 3 Proposed Service Allocation Technique in Cloud 3.1 Service (Task) Model 3.2 Proposed Algorithm 4 Simulations and Results 5 Conclusion References Small-Footprint Keyword Spotting in Smart Home IoT Devices 1 Introduction 2 Background 2.1 State of Art 2.2 Feature Extraction Techniques Explored in This Work 2.3 Neural Networks Explored in This Work 3 Training 3.1 Dataset and Experimental Setup 4 Deployment 4.1 Conversion of PyTorch Model to TensorFlow Lite 4.2 Open Neural Network Exchange (ONNX) 4.3 Post-training Quantization 4.4 Handling of Multiple Keywords 4.5 Inference in Raspberry Pi 5 Results 6 Conclusion References Overcoming an Evasion Attack on a CNN Model in the MIMO-OFDM Wireless Communication Channel 1 Introduction 2 CNN Implementation for a Wireless Communication Channel 2.1 Details of the Deep CNN Architecture 2.2 Performance Evaluation of the CNN Architecture 3 Evaluation of the Proposed Configuration to Variations of the Wireless Channel 3.1 Estimation of BERs 3.2 Automatic Modulation Classification 3.3 Physical Layer Security and Reliability 3.4 Success Rate of the Evasion Attack 4 Conclusion References An Improved Whale Optimization Algorithm for Optimal Placement of Edge Server 1 Introduction 2 Related Work 3 System Design 4 Proposed Improved Whale Optimization Algorithm-Based Edge Server Placement (ESP) 5 Experimental Results 5.1 Comparison of Delay 5.2 Comparison of Energy Consumption 5.3 Comparison of Convergence 6 Conclusion References Performance Analysis of LBT Cat4 Based 5G IoT Enabled New Radio in Unlicensed Spectrum 1 Introduction 1.1 NR-U Scenarios 1.2 Channel Access Categories 2 Coexistence Mechanism of Listen Before Talk (LBT) for NR-U 2.1 Modes of Transmission 2.2 Energy-Detection-Threshold 2.3 Contention Window-Size 2.4 Back-Off 2.5 Channel Access Priority Classes 3 Wi-Fi NR-U Co-Existence 3.1 Coexistence Performance Evaluation 3.2 Evaluation Methodology 3.3 Coexistence Performance 4 Conclusion References Lattice Cryptography-Based Geo-Encrypted Contact Tracing for Infection Detection 1 Introduction 2 Related Work 3 Background 4 Proposed Lattice Cryptosystem 4.1 Secret Key Generation Using Location Parameters 4.2 Generating Public Key 4.3 Encryption 4.4 Decryption 5 Performance Analysis 5.1 Time Complexity Analysis 5.2 Security Analysis 5.3 Possible Optimization 6 Conclusions References Beamforming Technique for Improving Physical Layer Security in an MIMO-OFDM Wireless Channel 1 Introduction 2 Theoretical and Methodological Backgrounds 3 Wireless Channel Model 4 Physical Layer Security and Success Rate of the Eve 5 Conclusion References Predictive VM Consolidation for Latency Sensitive Tasks in Heterogeneous Cloud 1 Introduction 2 Related Work 3 Problem Formulation 3.1 System Model 3.2 Problem Statement 4 Methodology 4.1 System Architecture 4.2 Resource Prediction 4.3 Energy Efficient Resource Management 4.4 Solution Approach 4.5 Non-slack Aware Approaches (Non-predictive) 4.6 Slack Aware Approaches (Predictive) 5 Experimental Evaluation 5.1 Experiment with Google Cluster Data 5.2 Result Analysis for Google Cluster Data 6 Conclusion and Future Work References Reporting Code Coverage at Requirement Phase Using SPIN Model Checker 1 Introduction 2 Related Work 3 Proposed Approach 4 Experimental Study 5 Conclusion References Metric-Oriented Comparison of Selective Forwarding Attack Detection Techniques in IoT-Based Systems 1 Introduction 2 Security Threats in IoT 2.1 Perception Layer Attacks 2.2 Communication Layer Attack ch14maria2020 2.3 Application Layer Attacks 3 Selective Forwarding Attack (SFA) 4 Literature Survey 5 Comparison and Analysis 6 Conclusion References Performance Enhancement of the Healthcare System Using Google Cloud Platform 1 Introduction 2 Literature Survey 3 Proposed Methodology 3.1 Collection of Data 3.2 Import Dataset to the Cloud Platform 3.3 Training Dataset 3.4 Note the Accuracy 3.5 Implementation of K-means Clustering Algorithm 3.6 Note the Increase in Accuracy 3.7 Implement of K-NN Classification Algorithm 3.8 Compare the Accuracy Level 4 Simulation Environment 5 Result and Discussions 6 Conclusion and Future Scope References Front-End Security Analysis for Cloud-Based Data Backup Application Using Cybersecurity Tools 1 Introduction 2 Related Works 3 Methodology of Experimentation 3.1 Tools Considered for Cyber Analysis 4 Results and Analysis 4.1 Discussion 5 Conclusions References Health Insurance Fraud Detection Using Feature Selection and Ensemble Machine Learning Techniques 1 Introduction 1.1 Categories of Health Insurance Frauds 1.2 Our Contributions 2 Related Work 3 Proposed Methodology 3.1 About the Dataset 3.2 Data Preprocessing 3.3 Algorithms and Models used for Model Building 3.4 Performance Metrics 4 Results and Observations 5 Conclusion and Future Scope References Real-Time American Sign Language Interpretation Using Deep Convolutional Neural Networks 1 Introduction 2 Related Works 3 Approach and Method 3.1 Network Architecture 3.2 Optimization and Loss Function 4 Data 4.1 Data Preprocessing 4.2 Data Augmentation 5 Evaluation and Model Analysis 6 Mobile Application 7 Conclusion References Multi-branch Multi-scale Attention Network for Facial Expression Recognition (FER) in-the-Wild 1 Introduction 2 Proposed Method 2.1 A Framework Overview 2.2 Multi-scale Module 2.3 Attention Module 2.4 Fusion Strategy and Loss Function 3 Results and Discussion 3.1 Datasets and Implementation Details 3.2 Comparison of Results 3.3 Ablation Analysis 4 Conclusion References Identifying COVID-19 Pandemic Stages Using Machine Learning 1 Introduction 2 Related Work 3 Pandemic Stage Identification 3.1 Three Pandemic Stages 3.2 Preparation of Dataset 3.3 Dataset Extraction 3.4 Machine Learning Algorithms 4 Results 5 Conclusion References A Multi-feature Analysis of Accented Multisyllabic Malayalam Words—a Low-Resourced Language 1 Introduction 2 Related Work 3 Proposed Methodology and Design 3.1 Dataset Construction 3.2 Experiment with MFCC 3.3 Experiment with STFT 3.4 Experiment with Combined MFCC and STFT 3.5 Experiment with Tempogram Features 3.6 Experiment with Tempogram, STFT and MFCC 3.7 Experiment with Tempogram and MFCC 3.8 Experiment Using MFCC, STFT, Mel Spectrogram, Spectral Roll off, Root Mean Square and Tempogram 4 Experimental Results 5 Conclusion and Future Scope References Machine Learning Based Fruit Detection System 1 Introduction 1.1 Related Work 2 Background Details 2.1 Real-Time Object Detection 2.2 Convolutional Neural Network (CNN) 2.3 YOLOv3 Algorithm 2.4 Architecture of YOLOv3 3 Proposed Methodology 4 Results and Discussions 5 Conclusion References Post hoc Interpretability: Review on New Frontiers of Interpretable AI 1 Introduction 2 Post hoc Interpretability on Tabular Data 3 Interpretable AI Toolkits 3.1 AIX360 3.2 Dr.Why.AI-Dalex 3.3 InterpretML 3.4 H2O Driverless AI 3.5 Amazon SageMaker Clarify 3.6 Quantus 4 Post hoc Interpretability Local Individual Predictions 4.1 LIME 4.2 Individual Conditional Expectation (ICE) 4.3 Counterfactual Explanations 4.4 Anchors 4.5 SHAP 5 Post hoc Explanation Vulnerabilities 6 Future Scope 7 Conclusion References SRGAN with 3D CNN Model for Video Stabilization 1 Introduction 2 Related Work 3 Proposed Model 3.1 Video Input 3.2 Extract Current Frames 3.3 Video Stabilization Model 3.4 Update Background Model 3.5 Background and Foreground Frames 3.6 Video Output 4 Experiment Result 4.1 Dataset 4.2 Parameter Setting 4.3 Discussion 5 Conclusion References Finding the Source of a Tweet and Analyzing the Sentiment of the User from h(is)er Tweet History 1 Introduction 2 Survey 3 Resources Used 3.1 Preparing a Knowledge Base 3.2 Lemmatization 3.3 English WordNet 4 Proposed Approach 5 Result and Corresponding Evaluation 6 Challenges 7 Conclusion and Future Scope of Work References Bitcoin Price Prediction by Applying Machine Learning Approaches 1 Introduction 1.1 Cryptocurrency 1.2 Blockchain 1.3 Theoretical Background of Bitcoin 2 Literature Study 3 Proposed Methodology 3.1 Collected Data 3.2 Data Pre-processing 3.3 Soft Computing Techniques Applied to the Prediction 3.4 Performance Evaluation 4 Experimental Result Discussion 4.1 Methodology Used 4.2 Result Analysis 5 Conclusion and Future Work References Social Engineering Attack Detection Using Machine Learning 1 Introduction 2 Literature Survey 3 Dataset Description 4 Implementation 5 Software and Libraries Used 6 Results and Discussions 7 Conclusion and Future Work References Corn Yield Prediction Using Crop Growth and Machine Learning Models 1 Introduction 2 Background 3 Methodology 3.1 Mechanistic Crop Growth Model 3.2 Machine Learning Models 4 Results 4.1 Nitrogen Application 4.2 ML Model Hyperparameter Configuration and Performance 4.3 Model Optimization 5 Conclusion References Deep Learning-Based Cancelable Biometric Recognition Using MobileNetV3Small Model 1 Introduction 2 Related Work 3 Proposed Methodology 3.1 MobileNet 3.2 Versions of MobileNet 3.3 MobileNetV3Small 3.4 Gaussian Random Projection 3.5 Random Forest 4 Experimental Results and Discussion 5 Conclusion and Future Work References Performance-Based Evaluation for Detection and Classification of Breast Cancer in Mammograms 1 Introduction 2 Materials and Methodologies 2.1 Datasets 2.2 Methodology 3 Results and Discussion 3.1 Performance Metrics and Results 4 Conclusions References Predictive Maintenance of NASA Turbofan Engines Using Traditional and Ensemble Machine Learning Techniques 1 Introduction 2 Methods and Materials 2.1 Dataset 2.2 Prediction Methodology 2.3 Algorithms 2.4 Performance Metrics 3 Result and Discussion 4 Conclusion and Future Work References Stacking a Novel Human Emotion Recognition Model Using Facial Features 1 Introduction 1.1 Design Challenges and Research Questions (RQs) 1.2 Contribution and Outline 2 Related Work 2.1 Facial Emotion Recognition Methods 3 Proposed Approach 4 Experimental Analysis 4.1 Data Settings 4.2 Human Emotion Recognition Accuracy Evaluation 5 Conclusion and Future Work References RecommenDiet: A System to Recommend a Dietary Regimen Using Facial Features 1 Introduction 2 Literature Survey 3 Methodology Overview 4 Results and Discussion 5 Conclusion References Image Transformation Based Detection of Breast Cancer Using Thermograms 1 Introduction 2 Related Work 3 Dataset Description 4 Proposed Methodology for Detection of Breast Cancer 4.1 Image Preprocessing 4.2 Image Transformation 4.3 Feature Extraction 4.4 Feature Ranking 5 Results 6 Conclusion References Vehicle Re-identification Using Convolutional Neural Networks 1 Introduction 2 Literature Survey 2.1 Related Work 2.2 Motivation 2.3 Problem Statement 3 Methodology 3.1 Datasets 3.2 Proposed Model 3.3 Impact of Filter Grafting 3.4 Semi-supervised Learning 3.5 Post-processing 4 Results and Analysis 4.1 Results 4.2 Model Performance 5 Conclusion and Future Work References Evaluation of Federated Learning Strategies on Industrial Time Series Fault Classification 1 Introduction 2 Overview of Existing FL Frameworks and Aggregation Strategies 2.1 Federated Learning Frameworks 2.2 FL Algorithms 3 Experiments and Evaluation with Industrial Data Set 3.1 Data Preparation 3.2 Lab Setup 3.3 Experiments and Observations 4 Conclusion References Optimized Algorithms for Quantum Machine Learning Circuits 1 Introduction 2 Classification 3 Different Approaches to Quantum Machine Learning 4 Dequantization 5 Polynomial Speedups 6 Learning Theory 6.1 No-Free Lunch Theorem 6.2 PAC Learning 7 Learning Quantum States 7.1 Classical Shadows 7.2 Hamiltonian Learning 7.3 Variational Quantum Circuits as QML Model 8 Quantum Optimization Algorithm and Loss Function 9 Conclusion and Future Work References Prediction of SOH and RUL for Lithium-Ion Batteries Using Regression Method with Feature of Indirect Related to SOH (FIRSOH) and Linear Time Series Model 1 Introduction 2 Experimental Information 2.1 Extraction of Feature of Indirect Related to SOH (FIRSOH) 3 Prediction of SOH Using LRR Method 4 RUL Prediction Using LTS Method 5 Conclusion References Chatbot for Mental Health Diagnosis Using NLP and Deep Learning 1 Introduction 2 Related Work 2.1 Proposed Method 2.2 Conversational Model 2.3 Classification Model 2.4 Response Generation 3 Experimental Analysis 4 Discussion 5 Conclusion and Future Scope References SincSquareNet: Deep Neural Network-Based Speaker Identification for Raw Speech 1 Introduction 2 The Sincsquarenet Architecture 3 Related Work 4 Experimental Setup & Results 4.1 Dataset 4.2 Experimental Configurations 4.3 Standard CNN 4.4 ConstantSincSquareNet 4.5 SincSquareNet 5 Conclusion References RSSI-Based Hybrid Approach for Range-Free Localization Using SA-PSO Optimization 1 Introduction 2 Related Work 2.1 Traditional “DV-Hop” Approach 2.2 RSSI 3 Improved Hybrid Optimization Technique (SAPSO) 3.1 Simulated Annealing (SA) Algorithm 3.2 The PSO Algorithm 4 Proposed SAPSODV-Hop 4.1 SAPSO-Based DV-Hop Positioning with RSSI 4.2 Optimized Node Localization Process Using SAPSO 5 Simulation Parameter and Experimental Environment 6 Conclusion References Multimodal Paddy Leaf Diseases Detection Using Feature Extraction and Machine Learning Techniques 1 Introduction 2 Related Works 3 Materials and Methods 3.1 Dataset 3.2 Pre-processing 3.3 Feature Extraction 3.4 Paddy Leaf Disease Classification 4 Results and Discussion 5 Conclusion References Quad Mount Fabricated Deep Fully Connected Neural Network Based Logistic Pricing Prediction 1 Introduction 2 Literature Review 3 Research Methodology 4 Implementation Setup and Results 5 Conclusion References Machine Learning and Deep Learning Models for Vegetable Leaf Image Classification Based on Data Augmentation 1 Introduction 2 Literature Review 3 Proposed Methodology 3.1 Dataset Preparation 3.2 Splitting Dataset into the Train, Validation, and Test Set 3.3 Using Deep Learning Model Training and Selections to Analyze Datasets 4 Experiments 5 Result and Analysis 5.1 Description of Results 5.2 Analysis of Results 6 Conclusion References Deep Fake Generation and Detection 1 Introduction 2 Related Work 2.1 Generic Overview 3 Proposed Methodology 3.1 Sample Dataset 3.2 The Architecture of the Proposed Method 3.3 First Order Motion 3.4 Long Short-Term Memory (LSTM) 3.5 Flow Diagram 3.6 Explanation of Algorithm 4 Results and Conclusion 5 Future Work References Similarity-Based Recommendation System Using K-Medoids Clustering 1 Introduction 2 Existing Work 3 Proposed System 4 Experimentation and Results 4.1 Online Clustering 4.2 Similarity Based Recommendations 5 Conclusion and Future Scope References Towards a General Black-Box Attack on Tabular Datasets 1 Introduction 2 Related Work 3 Background 3.1 Feature Importance Guided Attack (FIGA) 4 Datasets used for the Black-Box FIGA Attack 5 Methodology 5.1 Threat Model 5.2 Experimental Setup 5.3 Tuning FIGA 5.4 Evaluation Metrics 6 Results 7 Conclusion References Multi-task System for Multiple Languages Translation Using Transformers 1 Introduction 2 Related Works 3 Background and Methodology 3.1 Transformer Network 3.2 Proposed Approach 4 Experiments 5 Evaluation 6 Conclusion References Analysis of Various Hyperparameters for the Text Classification in Deep Neural Network 1 Introduction 2 Review of Literature 3 Research Gap 3.1 Numerous Layers 3.2 Numerous Neurons 3.3 Activation Function and its Features 3.4 Function Decay 3.5 Optimization Algorithms 3.6 Numerous Epochs 4 Methodology 5 Examining the Ideal Deep Learning Text Classification Environment 6 Conclusion References Analysis and Prediction of Datasets for Deep Learning: A Systematic Review 1 Introduction 2 Background 3 Datasets 3.1 Dataset Preprocessing and Model Implementation 4 Generalized Framework 5 Result Analysis 6 Conclusion References Lung Cancer Classification Using Capsule Network: A Novel Approach to Assist Radiologists in Diagnosis 1 Introduction 2 Related Works 3 Materials and Methods 4 Results and Discussions 5 Conclusion References Author Index