Advanced Computing: 11th International Conference, IACC 2021, Msida, Malta, December 18–19, 2021, Revised Selected Papers (Communications in Computer and Information Science, 1528)

Preface
Organization
Contents
Editors
Application of Artificial Intelligence and Machine Learning in Healthcare
Relating Design Thinking Framework in Predicting the Spread of COVID in Tamilnadu Using ARIMA
	Abstract
	1 Introduction
	2 Data Depiction
	3 Forecasting Models
	4 ARIMA Used for Forecasting COVID in Near Future
	5 Conclusion
	References
Covid Alert System: A Smart Security System to Alert Violations of Covid Protocol Using OpenCV
	Abstract
	1 Introduction
	2 Related Works
	3 Proposed Model
	4 Methodology
	5 Result
	6 Conclusion and Future Enhancement
	References
Corona Virus Detection Using EfficientNet from CT Scans
	Abstract
	1 Introduction
		1.1 Why CT Screenings?
		1.2 Existing Systems
		1.3 VGG-16
		1.4 ResNet-50
	2 Literature Review
	3 Deep Learning Models for Covid Detection
		3.1 EfficientNet B0 Architecture
		3.2 ResNet50 Architecture
	4 Experimental Results
	5 Conclusion
	References
Disease Diagnosis in Grapevines – A Hybrid Resnet-Jaya Approach
	1 Introduction
	2 Related Work
	3 Datasets Used
	4 Proposed Methodology
		4.1 Residual Networks (Resnet)
		4.2 Minimised Jaya Algorithm – Hybrid NN Approach
	5 Experimental Results and Analysis
		5.1 Computation Environment
		5.2 Optimum Learning Rate Selection
		5.3 Cyclical Annealing of the Learning Rate
		5.4 Handling Confident Inaccuracies
		5.5 Accuracy Analysis and Comparison with Previous Work
		5.6 Different CNN Model Comparisons
		5.7 Hybrid Jaya Algorithm Approach Compared with Other Optimisation Algorithms
	6 Conclusion
	References
Covid-19 Detection Using X-Ray Image
	Abstract
	1 Introduction
	2 Literature Review
	3 Methodology
		3.1 Flow Chart
		3.2 Phases of Model:
	4 Experimentation
	5 Results and Analysis
	6 Limitations
	7 Conclusion
	References
Alzheimer's Disease Classification Using Transfer Learning
	1 Introduction
		1.1 Related Work
		1.2 Organization
	2 Proposed Methodology
		2.1 Transfer Learning
		2.2 Convolutional Neural Networks
		2.3 VGG-19
		2.4 Inception-V3
		2.5 ResNet-50
		2.6 DenseNet-169
	3 Performance Evaluation Parameter
		3.1 Data Set
		3.2 Results and Discussion
	4 Conclusion
	References
A Comparative Study of Deep Learning Models for Detecting Pulmonary Embolism
	1 Introduction
	2 Materials
	3 Methodology
		3.1 Preprocessing
		3.2 Neural Network Module
		3.3 Training Parameters
		3.4 Models Used for Comparison
		3.5 Post Processing
	4 Results
	5 Conclusion
	6 Future Scope
	References
A Novel Compressed and Accelerated Convolution Neural Network for COVID-19 Disease Classification: A Genetic Algorithm Based Approach
	1 Introduction
	2 Materials and Methods
		2.1 Covid Disease Image Dataset
		2.2 Architecture of Convolution Neural Network for Disease Classification
	3 Results and Discussion
		3.1 Performance of ML Classifiers
	4 CNN Compression
	5 Conclusion
	References
Artificial Intelligence is Not a Foe But a Friend in the Healthcare Sector
	Abstract
	1 Introduction
	2 Review of Literature
	3 Research Methodology
	4 Analysis and Interpretation
		4.1 Discussion of the Model
	5 Conclusion and Recommendation
	6 Limitations and Scope for Future Study
	References
AI and The Cardiologist-When Mind, Heart and Machine Unite
	Abstract
	1 Introduction
	2 A Brief Introduction to AI Principles for the Clinician
		2.1 Machine Learning and Deep Learning
		2.2 Natural Language Processing
	3 Artificial Intelligence in Cardiology
		3.1 Electrophysiology
		3.2 Echocardiography
			3.2.1 Stress Echocardiography
		3.3 Clinical Decision Support and Preventive Cardiology
		3.4 Wearable Sensors
	4 Moving Forward: Future Prospects
	References
An Analysis of the Psychological Implications of COVID-19 Pandemic on Undergraduate Students and Efforts on Mitigation
	1 Introduction
	2 Literature Review
		2.1 SARS and Ebola Epidemics
		2.2 Psychological Effect of COVID-19 on General Population
		2.3 Psychological Effect of COVID-19 on Student Population
		2.4 Technology
	3 Methods
		3.1 Study Design and Participants
	4 Results and Discussion
		4.1 Dhriti - Multilingual Chatbot for Mental Health Support
		4.2 Discussion
		4.3 Best Practices and Limitations of the Study
	5 Conclusion and Future Work
	References
Network-Based Identification of Module Biomarker Associated with Hepatocellular Carcinoma
	Abstract
	1 Introduction
		1.1 Epidemiology
		1.2 Risk Factors
		1.3 Pathophysiology of HCC
		1.4 MicroRNA (miRNA) Biomarkers
		1.5 Systems Biology for Identification
	2 Materials and Methods
		2.1 TCGA miRNA-Seq Data Extraction and Differential Expression Analysis
		2.2 miRNA Co-expression Network Construction and Module Detection
		2.3 Overall Survival (OS) Analysis
	3 Results and Discussion
		3.1 TCGA miRNA-Seq Data Extraction and Differential Expression Analysis
		3.2 miRNA Co-expression Network Construction and Module Detection
		3.3 OS Analysis
	4 Conclusion
	References
Identifying Hub Nodes and Sub-networks from Cattle Rumen Microbiome Multilayer Networks
	Abstract
	1 Introduction
	2 Method
	3 Results
		3.1 Multilayer Network
		3.2 Node Hub Ranked Based on Topological Properties
		3.3 Biological Function Analysis
	4 Discussion
	5 Conclusion
	References
Application of AI for Emotion and Behaviour Prediction
Machine Learning-Based Psychology: A Study to Understand Cognitive Decision-Making
	Abstract
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Sample and Data Collection
		3.2 Data Analytics
		3.3 Technology Stack
	4 Results
		4.1 Case Studies
		4.2 Findings
		4.3 Visualiations
	5 Implication and Limitation
	6 Conclusion and Future Scope
	References
Predicting Stock Market Prices Using Sentiment Analysis of News Articles
	Abstract
	1 Introduction
	2 Related Work
	3 Proposed Method
	4 Description of Data Set and Its Pre-processing
		4.1 Data Collection
		4.2 Data Processing
		4.3 Stock Prediction
	5 Results
	6 Conclusion
	References
Sentimental Analysis on Multi-domain Sentiment Dataset Using SVM and Naive Bayes Algorithm
	Abstract
	1 Introduction
	2 Related Works
	3 Proposed Methodolgy
		3.1 Naive Bayes Approach
		3.2 SVM
	4 Results and Discussion
		4.1 Accuracy Classification
		4.2 Confusion Matrix
	5 Conclusion
	References
Artificial Intelligence in Online Stores' Processes
	1 Introduction
	2 AI in Online Marketing
		2.1 Facebook Advertising
		2.2 Amazon Advertising
		2.3 Google - Smart Ads
	3 AI in an e-Commerce Fulfillment Center
	4 AI in Barcode Identification
	5 New AI-Driven Delivery Methods in E-Commerce
		5.1 Autonomous Delivery Robots
		5.2 Deliveries by Aerial Drones
		5.3 Deliveries Using 3D Printing
	6 Conclusions and Discussion
	References
Early Warning Indicators for Financial Crisis During Covid-19
	1 Introduction
	2 Methodology
		2.1 Detrending
		2.2 Early Warning Signal (EWS) Indicators
	3 Results and Discussion
	4 Conclusion
	References
Facial Recognition Based Attendance Monitoring System
	1 Introduction
	2 Literature Study
	3 Methodology
		3.1 Framework
	4 Design Approach
		4.1 Module 1
		4.2 Module 2
		4.3 Module 3
		4.4 Module 4
		4.5 Module 5
		4.6 Module 6
	5 Results and Analysis
	6 Conclusion and Future Work
	References
Transfer Learning Using Variational Quantum Circuit
	1 Introduction
	2 Background
	3 Related Works
	4 The Hybrid Model
		4.1 Transfer Learning
		4.2 Proposed Model
		4.3 Dressed Quantum Circuit
	5 Testing
	6 Results
	7 Discussion
	8 Future Work
	References
Problem Solving Using Reinforcement Learning and Analysis of Data
Gait Learning Using Reinforcement Learning
	1 Introduction
	2 Related Work
	3 Design and Learning Process
	4 Interfacing with the Real World Quadruped
	5 Results of Simulation
	6 Discussion
	7 Conclusion and Future Work
	References
Data Science in the Business Environment: Architecture, Process and Tools
	Abstract
	1 Introduction
	2 Architecture and Framework
		2.1 Data Science Conceptual Architecture
		2.2 EDISON Data Science Framework
		2.3 Case Study: Data Science Degree Apprenticeship
	3 Methodology and Process
		3.1 Range of Approaches
		3.2 A Process Model for Data-Driven Decision Making
		3.3 Case Study: Customer Analytics in SMEs
	4 Tools and Techniques
		4.1 EDISON Data Science Skills
		4.2 Analytical Technology
		4.3 Case Study: Marketing Analytics for Health Insurance
			4.3.1 Explore Data
			4.3.2 Discover Insights
	5 Conclusion
	References
A Logarithmic Distance-Based Multi-Objective Genetic Programming Approach for Classification of Imbalanced Data
	1 Introduction
	2 Materials and Methods
		2.1 Multi-Objective Genetic Programming (MOGP)
		2.2 Proposed MOGP Approach
		2.3 Trapezoidal Rule for Area Under Generated Pareto-Front
		2.4 GP Parameter Setting
		2.5 Benchmark Data Set
	3 Results and Discussion
	4 Conclusions
	References
Multiview Classification with Missing-Views Through Adversarial Representation and Inductive Transfer Learning
	1 Introduction
		1.1 Motivation and Contributions
	2 Related Work
	3 Proposed Methodology
		3.1 Adversarial Cross-View Latent Learning
		3.2 Classification Through an Inductive Transfer of Encodings
	4 Experimental Setup
		4.1 Datasets
		4.2 Evaluation Metrics
	5 Results Analysis and Discussion
		5.1 Performance on Existing Multiview State-of-the-Art Systems
		5.2 Performance Evaluation on Handling the Missing-Views in Multiview Data
	6 Conclusion and Future Perspectives
	References
Deep Reinforcement Learning Based Throughput Maximization Scheme for D2D Users Underlaying NOMA-Enabled Cellular Network
	1 Introduction
		1.1 Related Work
		1.2 Motivation and Contributions
		1.3 Organization
	2 System Model and Problem Formulation
		2.1 Channel Model
		2.2 Throughput Calculation
		2.3 Problem Formulation
	3 Proposed Solution
		3.1 Centralised Optimisation
		3.2 Proximal Policy Optimization
	4 Performance Evaluation
		4.1 Numerical Settings
		4.2 Results and Discussion
	5 Conclusion
	References
An Intrusion Detection System for Blackhole Attack Detection and Isolation in RPL Based IoT Using ANN
	1 Introduction
	2 Black Hole Attack in RPL Scenario
	3 Related Work
	4 INSULATE - The Proposed Intrusion Detection System
		4.1 System Model
		4.2 Local Detection
		4.3 Global Detection and Eviction
	5 Performance Evaluation
		5.1 Simulation Setup and Performance Metrics
		5.2 Simulation Results and Performance Comparison
	6 Conclusion
	References
Evaluating the Efficacy of Different Neural Network Deep Reinforcement Algorithms in Complex Search-and-Retrieve Virtual Simulations
	Abstract
	1 Introduction
	2 Background
		2.1 Proximal-Policy Optimization (PPO)
		2.2 Soft-Actor Critic (SAC)
	3 Methods
		3.1 Experimental Design
	4 Performance Metric
	5 Results
		5.1 Training Results
		5.2 Test Results
		5.3 Influence of Number of Hidden Layers of the Neural Network on Model Performance
		5.4 Influence of Number of Nodes of the Neural Network on Model Performance
	6 Discussion
	7 Future Scope
	8 Conclusion
	Acknowledgements
	References
Post-hoc Explainable Reinforcement Learning Using Probabilistic Graphical Models
	1 Introduction
	2 Related Work
	3 Methods
		3.1 Reinforcement Learning (RL)
		3.2 Explainable Reinforcement Learning (XRL)
		3.3 Probabilistic Graphical Model (PGM)
	4 Explainability and Reinforcement Learning
	5 Experimental Setup
	6 Results and Discussions
		6.1 Environment I : Taxi-V2
		6.2 Environment II: FrozenLake-V0
		6.3 Discussion
	7 Limitations and Future Work
	8 Conclusion
	References
Ghostbusters: How the Absence of Class Pairs in Multi-Class Multi-Label Datasets Impacts Classifier Accuracy
	1 Introduction
	2 Related Work
	3 Experimental Setup
	4 Experiments
		4.1 Experiment 1: Impact of Ratio Bias
		4.2 Experiment 2: Impact of Parity Bias
		4.3 Experiment 3: Impact of Degree of Separation
		4.4 Experiment 4: Adding Connected Examples to the MCML Classifier
	5 Discussion
	6 Future Work
	7 Conclusion
	8  Appendix
	References
ReLearner: A Reinforcement Learning-Based Self Driving Car Model Using Gym Environment
	1 Introduction
	2 Related Works
	3 ReLearner Model
		3.1 No Reinforcement Learning
		3.2 Q-learning
	4 Experimental Results
	5 Conclusion
	References
Automating Paid Parking System Using IoT Technology
	Abstract
	1 Introduction
	2 Literature Review
	3 Architectural Diagram
	4 Proposed Model
		4.1 Classification of Vehicles
		4.2 Real Time Number Plate Detection
		4.3 Character Segmentation and Recognition
		4.4 IOT Using RPi
		4.5 PayPark Application
	5 Observations and Results
	6 Conclusion
	7 Future Scope
	References
Farmers' Survey App - An Interactive Open-Source Application for Agricultural Survey
	1 Introduction
	2 Key Features of the Android Application
	3 Application Interface and Architecture of Survey App
		3.1 Application Interface
		3.2 The Architecture of the Survey App
	4 Case Study: Analysis of Survey Data
	5 Conclusion
	References
Advance Uses of RNN and Regression Techniques
Improving Recognition of Handwritten Kannada Characters Using Mixup Regularization
	Abstract
	1 Introduction
	2 Literature Survey
	3 Dataset
		3.1 Kannada84
		3.2 Char74K
	4 Proposed Methodology
		4.1 Pre-processing
		4.2 Augmentation
	5 Architecture
		5.1 VGG Network
		5.2 Residual Neural Network
		5.3 Squeeze-and-Excitation Networks
		5.4 Improving Performance with Mixup
	6 Training
	7 Results
	8 Conclusion and Future Work
	References
Research on the Detection and Recognition Algorithm of Click Chinese Character Verification Code
	Abstract
	1 Introduction
	2 Solution Idea
	3 Detection Experiment
		3.1 K-means Clustering Generates Anchor Size
		3.2 Class Imbalance
	4 Recognition Experiment
		4.1 Main Idea
		4.2 Feature Extraction Network
		4.3 Difference of Gaussian
		4.4 Similarity Calculation
	5 Conclusion
	References
Multihead Self-attention and LSTM for Spacecraft Telemetry Anomaly Detection
	1 Introduction
	2 LSTM Model for Anomaly Detection
	3 MHSA Model for Anomaly Detection
		3.1 Positional Encoding
		3.2 Architecture
	4 Non-parametric Dynamic Thresholding (NPDT) Method
	5 Results and Analysis
		5.1 Detection of Solar Array Drive Assembly (SADA) Anomaly
		5.2 Detection of Momentum Wheel Anomaly
		5.3 Detection of Reaction Wheel Anomaly
	6 Comparison of the Models
		6.1 Number of Parameters
		6.2 Empirical Time Complexity of the Models
		6.3 Comparing Normalized Anomaly Scores
	7 Discussions and Conclusions
	References
Validity and Reliability Assessment of a Smartphone Application for Measuring Chronic Low Back Pain
	Abstract
	1 Introduction
	2 Related Work
		2.1 Chronic Pain: More Than Just a Physical Problem
		2.2 Chronic Low Back Pain (CLBP): Definition and Epidemiology
		2.3 Mode of Assessment
	3 Problem Formulation
	4 Methodology
		4.1 Unit Study Design
		4.2 Tools
		4.3 Testing Procedure
	5 Results and Discussions
		5.1 Participants
		5.2 Readings
		5.3 Analysis of Data
		5.4 Discussions
		5.5 Limitations
	6 Conclusions and Future Scope
	References
Predicting Disasters from Tweets Using GloVe Embeddings and BERT Layer Classification
	Abstract
	1 Introduction
	2 Literature Survey
		2.1 Text Classification Using Machine Learning Techniques
		2.2 Sentimental Analysis of Twitter Data Using Classifier Algorithms
		2.3 Event Classification and Retrieving User’s Geographical Location Based on Live Tweets on Twitter and Prioritizing Them to Alert the Concern Authority
		2.4 Automatic Classification of Disaster-Related Tweets
		2.5 Comparing BERT Against Traditional Machine Learning Text Classification
		2.6 Usage and Analysis of Twitter During 2015 Chennai Flood Towards Disaster Management
		2.7 A Comparative Analysis of Machine Learning Techniques for Disaster-Related Tweet Classification
		2.8 Multimodal Analysis of Disaster Tweets
	3 Methodology
		3.1 Data Exploration
		3.2 Meta Features Extraction
		3.3 Text Cleaning and Preprocessing
		3.4 Cardinality and Target Distribution
		3.5 Target Features
		3.6 Architecture of the Disaster Detector Function
		3.7 Word Embedding Using GloVe
		3.8 Classification Using BERT Model
	4 Results
	5 Conclusion
	References
Contextual Quality Assessment of the Newspaper Articles Based on Keyword Extraction
	1 Introduction
	2 Related Works and Our Contributions
	3 Preliminaries
		3.1 Quality of Document
		3.2 Graph Centrlity Metrics
		3.3 Clustering Coefficient
		3.4 Mantel Test
	4 Our Approach for Quality Detection of Any Article
		4.1 Representation of the Text Report with Graph
		4.2 Analysing the Generated Graph
		4.3 Features
		4.4 Illustration with an Example
	5 Result and Analysis
		5.1 Comparison with Other Existing Methods
		5.2 Result on Real-Life Application
		5.3 Result on Newspaper Data
		5.4 Limitations
	6 Conclusions
	References
GEDset: Automatic Dataset Builder for Machine Translation System with Specific Reference to Gujarati-English
	Abstract
	1 Introduction
	2 Motivation and Significance
	3 Software Description
		3.1 Software Architecture
		3.2 Software Functionalities
	4 Proposed Algorithm
	5 Experiment and Result
	6 Conclusion
	References
Power Function Algorithm for Linear Regression Weights with Weibull Data Analysis
	Abstract
	1 Introduction
	2 Asymptotic Power Functions
	3 Weibull Parameter Estimation by WLR
		3.1 Weighted Linear Regression
		3.2 Variances of the Plotting Position Z
	4 Power Model for LR Weights
		4.1 Infinite Sample Size
		4.2 Finite Sample Size
	5 Evaluation of the Algorithm
	6 Discussion and Conclusion
	References
Special Intervention of AI
Shrub Detection in High-Resolution Imagery: A Comparative Study of Two Deep Learning Approaches
	1 Introduction
	2 Approach
		2.1 Dataset and Target Plant
		2.2 CNN-Based Detection Approach
		2.3 Segmentation Approach
		2.4 Evaluation
	3 Experimental Evaluation
		3.1 Classification of Ideal Image Patches
		3.2 Detection – Classification of Proposed Windows
		3.3 Deep Learning-Based Semantic Segmentation
		3.4 Comparative Results
	4 Discussion
	5 Conclusion
	References
Optimized Deep Neural Network for Tomato Leaf Diseases Identification
	Abstract
	1 Introduction
	2 Optimized Deep Neural Network for Tomato Leaf Diseases (ODNN-TLD)
		2.1 Data Preparation
		2.2 Data Pre-processing
		2.3 Training the Proposed Model with VGG16 and CNN (ODNN-TLD)
		2.4 Optimizing Functions
	3 Results and Discussion
	4 Conclusion
	Acknowledgement
	References
Application of Distributed Back Propagation Neural Network for Dynamic Real-Time Bidding
	1 Introduction
	2 Background
		2.1 Real-Time Bidding
		2.2 Distributed Back Propagation Neural Networks (d-bpnn)
	3 Working of Bidding Strategies
		3.1 Flat Bidding
		3.2 Randomized Bidding
		3.3 ML Bidding
	4 Empirical Evaluation and Discussion
		4.1 Parameters of Evaluation
		4.2 Description of Dataset
		4.3 Experimental Setup and Results
	5 Conclusions
	References
An Efficient Minimum Spanning Tree-Based Color Image Segmentation Approach
	Abstract
	1 Introduction
	2 Related Works
	3 Proposed Methodology
	4 Experimental Setup
		4.1 Dataset Used
		4.2 Evaluation Parameters
	5 Experimental Results and Discussion
	6 Conclusion and Future Perspective
	References
Geo-ML Enabled Above Ground Biomass and Carbon Estimation for Urban Forests
	Abstract
	1 Introduction
	2 Materials and Methods
		2.1 Study Area
		2.2 Satellite Data
		2.3 Field Sampling
		2.4 Field Observed AGB and Carbon Estimation for Urban Forest
	3 Methods
		3.1 Geospatial Modelling of AGB and Carbon
		3.2 Variable Selection
		3.3 RF Based Approach
	4 Results and Discussion
		4.1 Ground Observed AGB and Carbon Estimation
		4.2 AGB Estimation Using RF
		4.3 Spatial Distribution of AGB and Carbon
	5 Conclusion
	References
Custom Cloud: An Efficient Model for Cloud Service Selection Based on Neural Network
	Abstract
	1 Introduction
		1.1 Motivation and Contributions
		1.2 Paper Outline
	2 Proposed Custom Cloud Model
	3 Case Study
	4 Experimental Results and Discussion
	5 Conclusion
	Data and Code Availability
	References
A Machine Learning-Based Approach for Efficient Cloud Service Selection
	Abstract
	1 Introduction
		1.1 Paper Outline
	2 Literature Survey
	3 Problem Statement
	4 Experimental Results and Discussion
	5 Conclusion
	References
Enhancing Network Robustness Using Statistical Approach Based Rewiring Strategy
	1 Introduction
	2 Literature Review
	3 Definitions
	4 Proposed Strategies
		4.1 Betweenness Centrality and Core Periphery (BCCP) Based Rewiring Strategy
		4.2 Clustering Coefficient and Betweenness Centrality (CCBC) Based Rewiring Strategy
	5 Results
	6 Conclusions
	References
A Partcle Swarm Optimization Based Approach for Filter Pruning in Convolution Neural Network for Tomato Leaf Disease Classification
	1 Introduction
	2 Materials and Methods
		2.1 Dataset
		2.2 Proposed Model
	3 Experimental Results
		3.1 Pre-trained CNN Models Performance
		3.2 Performance of Machine Learning Methods
	4 CNN Model Compression
	5 Conclusion
	References
Data Breach in Social Networks Using Machine Learning
	Abstract
	1 Introduction
	2 Literature Review
	3 Theoretical Background
		3.1 Breach Methods
		3.2 What Causes Data Breaches?
		3.3 Recent Data Breaches & Statistics
		3.4 Data Analysis
	4 Results
	5 Conclusion
	References
Sentiment Analysis of Customers Review Using Hybrid Approach
	Abstract
	1 Introduction
		1.1 Natural Language Processing
		1.2 Sentiment Analysis
		1.3 Objective of Sentiment Analysis
		1.4 Sentiment Analysis Approach for Prediction
		1.5 Different Types of Sentiment Analysis
	2 Literature Review
	3 Problem Statement
	4 Proposed Model
		4.1 Objective of Proposed Work
		4.2 Working of Proposed Work
	5 Result and Discussion
		5.1 Accuracy Comparison
		5.2 Performance Comparison
	6 Conclusion
	7 Scope of Research
	References
Author Index