Preface
Organization
Program Committee
Chairman
Vice-chairmen
Members
Organization Committee
Chairman
Vice-chairmen
Members
Contents
Analysis and Modeling of Hybrid Systems and Processes
Application of Convolutional Neural Network for Gene Expression Data Classification
1 Introduction
2 Literature Review
3 Material and Methods
3.1 Architecture, Structure and Model of Convolutional Neural Network
4 Simulation, Results and Discussion
4.1 Gene Expression Dataset Formation and Preprocessing
4.2 Application of 1D One-Layer CNN for Gene Expression Data Classification
4.3 Application of 1D Two-Layer CNN for Gene Expression Data Classification
4.4 Model of 2D Convolutional Neural Network
4.5 Model of 2D Three-Layer Convolutional Neural Network
4.6 Estimation of CNN Robustness to Different Levels of Noise Component
5 Conclusions
References
Formation of Subsets of Co-expressed Gene Expression Profiles Based on Joint Use of Fuzzy Inference System, Statistical Criteria and Shannon Entropy
1 Introduction
2 Problem Statement
3 Literature Review
4 Fuzzy Model of Removing the Non-informative Gene Expression Profiles by Statistical and Entropy Criteria
4.1 Simulation Regarding Practical Implementation of the Proposed Fuzzy Logic Inference Model
5 Assessing the Fuzzy Inference Model Adequacy by Applying the Gene Expression Data Classification Technique
6 Conclusions
References
Mathematical Model of Preparing Process of Bulk Cargo for Transportation by Vessel
1 Introduction
2 Problem Statement
3 Literature Review
4 Materials and Methods
5 Experiment and Results
6 Discussions
7 Conclusions
References
Computer Simulation of Joule-Thomson Effect Based on the Use of Real Gases
1 Introduction
2 Literature Review
3 Materials and Methods
3.1 Theoretical Describing the Joule-Thomson Effect
3.2 Calculation of Heating System Efficiency Based on the Joule-Thomson Effect
3.3 Calculation of Heating System Efficiency Whose Working Fluid is Water
4 Simulation, Results and Discussion
5 Conclusions
References
Simulating Soil Organic Carbon Turnover with a Layered Model and Improved Moisture and Temperature Impacts
1 Introduction
2 Literature Review
3 Mathematical Model
3.1 Layered SOC Decomposition Model
3.2 Soil Moisture and Temperature Model
3.3 Abiotic Stress Functions
4 Experiment
4.1 Experimental Setting
4.2 Data Sources
5 Results and Discussion
6 Conclusions
References
Optimization of Coagulant Dosing Process for Water Purification Based on Artificial Neural Networks
1 Introduction
2 Problem Statement
3 Literature Review
4 Materials and Methods
4.1 Water Purification Process
4.2 Determination of Coagulant Dose
4.3 Modeling of Artificial Neural Network
5 Experiment, Results and Discussion
6 Conclusions
References
Methodology for Solving Forecasting Problems Based on Machine Learning Methods
1 Introduction
2 Forecasting Methodology
3 Implementation of the Methodology
3.1 Implementation of Regression Models
3.2 Implementation of Models Based on Trees
3.3 Implementation of Neural Network Models
3.4 Discussion
4 Conclusions
References
The Comprehensive Model of Using In-Depth Consolidated Multimodal Learning to Study Trading Strategies in the Securities Market
1 Introduction
2 Literature Review
3 Materials and Methods
3.1 Market Indicators
3.2 Market Model
3.3 Environment
3.4 Sampling Rate
3.5 Depth Neural Network
3.6 Q-learning
3.7 Reinforcement Learning
3.8 Actor-critic Model
3.9 Advantage Actor Critic (A2C)
3.10 Deep Deterministic Policy Gradient (DDPG)
3.11 Proximal Policy Optimization (PPO)
3.12 Ensemble of Models
4 Experiment and Results
4.1 Training of Individual Models of the Ensemble
4.2 Training and Analysis of the Ensemble
4.3 Behavior of the Ensemble Model with High Market Turbulence
5 Discussion
6 Conclusions
References
Mathematical and Computer Model of the Tree Crown Ignition Process from a Mobile Grassroots Fire
1 Introduction
2 Literature Review
3 Problem Statement
4 Matherial and Methods
5 Experiment, Results and Discussion
6 Conclusions
References
Features of Complex Application of the Formal Method of EVENT-B for Development of Environmental Management Systems
1 Introduction
2 Review of Literature
3 Material and Methods
3.1 Features of Application of the Method of Formal Verification Model Checking
3.2 The Essence and Basic Principles of the Event-B Requirements Specification Method
3.3 The Essence and Basic Principles of Constructing FTA Failure Trees
3.4 The Essence and Basic Principles of the Method of Analysis of Functional Stability of FME (C) A
3.5 FMECA-Analysis of the Event-B Model
3.6 The Method of Alternating Parameter Changes (Gauss-Seidel Method) in the Environmental Management System
3.7 The Method of Random Blind Searches in the System of Ecological Management
4 Computer Simulation, Results and Discussion
5 Conclusions
References
Ecology Objects Recognition by Optimized Inverse Filtration of Textured Background
1 Introduction
2 Literature Review
3 Problem Statement
4 The Method of Inverse Filtration Based on the Eigen Harmonic Decomposition
4.1 Object Recognition on Textured Background as Problem of Optimal Filtration
4.2 Eigen Harmonic Decomposition of Textured Image
4.3 Development of IRF for Texture Recognition
4.4 Optimization of the IRF
4.5 Eigen Harmonic Decomposition for the IRF Design
5 Experiment
6 Results and Discussion
7 Conclusions
References
Theoretical and Applied Aspects of Decision-Making Systems
Information Technology to Assess the Enterprises' Readiness for Innovative Transformations Using Markov Chains
1 Introduction
2 Related Works
3 Materials and Methods
4 Experiment, Results and Discussion
5 Conclusions
References
Method to Find the Original Source of COVID-19 by Genome Sequence and Probability of Electron Capture
1 Introduction
2 Material and Methods
2.1 Comparing Genome Sequences
2.2 Calculating the Probability of the Electron Capture
3 Results
4 Conclusions
References
Leader-Follower Strategy of Fixed-Wing Unmanned Aerial Vehicles via Split Rejoin Maneuvers
1 Introduction
2 Vehicle Model
3 Artificial Potential Field Function
3.1 Attractive Potential Field Functions
3.2 Repulsive Potential Field Functions
4 Design of the Acceleration Controllers
4.1 Lyapunov Function
4.2 Nonlinear Acceleration Controllers
5 Stability Analysis
6 Simulation Results and Discussion
6.1 Scenario 1: Arrowhead Formation in the Presence of Obstacles
6.2 Scenario 2: Double Platoon Formation in the Presence of Obstacles
7 Conclusion
References
Prognostic Assessment of COVID-19 Vaccination Levels
1 Introduction
2 Problem Statement
3 Literature Review
4 Material and Methods
5 Experiment
6 Results and Discussion
7 Conclusions
References
Application of the Theory of Functional Stability in the Problems of Covering Territories by Sensory Networks
1 Introduction
2 Related Works
3 Material and Methods
3.1 Criteria and Indicators of Functional Stability of a Complex System
3.2 Sustainability of Information and Telecommunication Systems in Emergency Situations
3.3 Technique to Ensure the Functional Stability of the Coverage of Territories by Sensor Networks
3.4 Computer Simulation, Results and Discussion
4 Conclusions
References
Adaptive Decision-Making Strategies in the Game with Environment
1 Introduction
2 Related Works
3 Materials and Methods
4 Experiments and Results
5 Discussion
6 Conclusions
References
System Analysis of the Internal and External Migration Processes in Ukraine
1 Introduction
2 Literature Review
3 Problem Statement
4 Materials and Methods
4.1 Statistical Analysis of the Population in Ukraine
4.2 Analysis of Internal Migration in Ukraine
4.3 Analysis of External Migration in Ukraine
5 Results and Discussions
6 Conclusions
References
Associative Information Retrieval in Medical Databases
1 Introduction and Literature Review
2 Problem Statement
3 Materials and Methods
3.1 Method and Basic Assumptions
3.2 Development of an Associative Information Retrieval Method Based on Determining the Degree of Similarity of Two Strings
3.3 Development of Criteria for the Similarity of Text Strings Based on the Selected Method
4 Experiment, Results and Discussion
5 Conclusions and Future Work
References
Analysis of Deep Learning Methods in Adaptation to the Small Data Problem Solving
1 Introduction
2 Related Works and Problem Statement
3 Data Preparation in Machine Learning and Deep Learning
3.1 General View on the Pipeline of Data Preparation
3.2 Detection of Anomalies in Data
3.3 Data Augmentation
3.4 Perturbation Compensation and Stability of Classification Algorithms
3.5 Orthogonal Transformations in Machine and Deep Learning
4 Note on Implementation of Algorithms for the Task
4.1 Overview of Libraries for Machine Learning Used in Study
4.2 Implementations for Processors
5 Evaluation of the Practical Effectiveness of Deep Learning Methods on Real Data
5.1 Data Used in Experiments and Previous Study in Area
5.2 Application of Computer Vision Techniques for Neural Networks
5.3 Latent Representation of Features as an Alternative to Data Dimensionality Reduction Methods
5.4 Performance of Deep Learning Methods on Small and Large Datasets
6 Conclusions
References
Cognitive and Information Decision Support Technologies for Operational Management of Energy-Active Objects in Boundary Modes
1 Introduction
2 Problem Statement
3 Literature Review
4 Materials and Methods
4.1 Analysis of Problematic Tasks of Selection and Cognitive Processing of Data in Making Target Decisions on the Control of the ACS Operator
4.2 Analysis of the Level of System and Cognitive Risks in Decision-Making in a Complex Technogenic System Under Conditions of Active Factors of Threats and Information Attacks
5 Experiment, Results and Discussion
6 Conclusions
References
Expert Decision Support System Modeling in Lifecycle Management of Specialized Software
1 Introduction
2 Problem Statement
3 Literature Review
4 Materials and Methods
5 Experiment, Results and Discussion
6 Conclusions
References
Data Engineering, Computational Intelligence and Inductive Modeling
Machine Learning of the Biotechnic System for Gastroesophageal Reflux Disease Monitoring
1 Introduction
2 Review of Literature
3 Problem Statement
4 Material and Methods
4.1 Data
4.2 Experimental Technique
4.3 Experiment Reliability
5 Experiment
5.1 Monitoring and Measuring
5.2 Binding Between Variables and Data Validity
6 Discussion
7 Conclusions
References
Processing Technology of Thematic Identification and Classification of Objects in the Multispectral Remote Sensing Imagery
1 Introduction
2 Related Works
3 Proposed Technology of Thematic Classification and Identification of Objects in the Remote Sensing Imagery
3.1 Pre-processing of Remote Sensing Data
3.2 Segmentation of Remote Sensing Imagery
3.3 Classification
4 Classification Assessment
5 Conclusions
References
CTrace: Language for Definition of Epidemiological Models with Contact-Tracing Transmission
1 Introduction
2 Literature Review
3 Method
3.1 Contract Tracing Model Overview
3.2 Model Definition
4 Modeling Language
4.1 Translator Intrinsics
4.2 Expressions
4.3 Parameters
4.4 Distributions
4.5 Functions
4.6 Arrays Structure
4.7 Compartments and Agents
4.8 Disease Model
5 Experiment, Results and Discussion
6 Conclusions
A Appendix
References
Optimization of Data Preprocessing Procedure in the Systems of High Dimensional Data Clustering
1 Introduction
2 Problem Statement
3 Literature Review
4 Material and Methods
5 Experiment, Results and Discussion
6 Conclusions
References
Features of the Application of the Principal Component Method to the Study of Acoustic Emission Signals Under Loading of Multilayer Structures
1 Introduction
2 Literature Review
3 Methodology
4 Materials and Methods
5 Experiment, Results and Discussion
6 Conclusions
References
Computational Intelligence in Medicine
1 Introduction
2 Problem Statement
3 Literature Review
4 Immunohistochemical Image Analysis
4.1 Image Segmentation
4.2 Software Structure Module for Automatic Image Segmentation
4.3 U-net Architecture for Immunohistochemical Image Segmentation
5 Computer Experiments
6 Image Generation
7 Cytological Image Classification
8 Conclusions
References
Approaches and Techniques to Improve Machine Learning Performance in Distributed Transducer Networks
1 Introduction
2 Literature Review
3 Materials and Methods
4 Experiment
5 Results and Discussion
6 Conclusions
References
Investigation of the Impact of Primary Data Processing on the Results of Neural Network Training for Satellite Imagery Recognition
1 Introduction
2 Literature Review
3 Problem Statement
4 Materials and Methods
4.1 Convolutions
4.2 Strides
4.3 Learning and Activation
4.4 Downsampling
4.5 Loss Function
4.6 Dataset Preparation
4.7 Neural Network Architecture
5 Experiment, Results and Discussion
5.1 Experiment 1: Raw Data for Training and Validation
5.2 Experiment 2: Raw Data for Training and Masked for Validation
5.3 Experiment 3: Masked Data for Training and Raw for Validation
5.4 Experiment 4: Masked Data for Training and Validation
6 Conclusions
References
Application of Wavelet Transform for Machine Learning Classification of Time Series
1 Introduction
2 Literature Survey
3 Materials and Methods
3.1 Continuous Wavelet Transform in Time Series
3.2 Input Data
3.3 Classification Metrics
4 Experiment
5 Results and Discussion
5.1 Time Realizations and Wavelet Spectra
5.2 Classification of Partially Noisy Sinusoids
6 Conclusions
References
A Noise Resistant Credibilistic Fuzzy Clustering Algorithm on a Unit Hypersphere with Illustrations Using Expression Data
1 Introduction
2 Related Works
3 Background
3.1 Fuzzy C-Means (FCM)
3.2 Hyperspherical Fuzzy C-Means (HFCM)
3.3 Credibilistic Fuzzy C-Means (CFCM)
4 Credibilistic Fuzzy C-Means-Direction (CFCM-D)
4.1 Standardization of the Feature Vectors on a Unit Hypersphere
4.2 Method
5 Performance Metrics
5.1 Precision and Recall
5.2 Silhouette Coefficient (SC)
5.3 Davies-Bouldin (DB) Score
5.4 Calinski-Harabasz (CH) Score
6 Results and Discussion
7 Conclusion
References
Visual Analytics-Based Method for Sentiment Analysis of COVID-19 Ukrainian Tweets
1 Introduction
2 Literature Review
3 Problem Statement
4 Materials and Methods
4.1 Preparation of Experimental Data
4.2 Construction of Vector of Features (Model) for Linear Classification by Visual Analytics
4.3 Assessment of the Classifier Quality
4.4 Experiment
5 Results and Discussion
6 Conclusions
References
Software Based on Ontological Tasks Models
1 Introduction
2 Related Works
3 Materials and Methods
3.1 The Use of Task Ontologies in Software Design
3.2 Formal Representation of Knowledge and Modeling System Structure
3.3 Mathematical Model of Representation and Processing of Knowledge in the Modeling System
3.4 The Model of Ontological Task Models Interactions
4 Experiment, Results and Discussion
4.1 Software as a Multi-leveled Hybrid System Based on Ontological Task Models
4.2 The Use of Ontological Task Models in Software. Application Example
4.3 The Modeling Environment for Ontological Models
4.4 Analysis of Advantages Using Ontological Models in Software Design
5 Conclusions
References
Neural Network Analysis of Evacuation Flows According to Video Surveillance Cameras
1 Introduction
2 Problem Statement
3 Literature Review
4 Stepwise Procedure of the Evacuation's Parameters Determining Optimisation
5 Experiment, Results and Discussion
6 Conclusions
References
Real-Time Information Technology Human Detection Using Cloud Services
1 Introduction
2 Materials and Methods
3 Experiment
4 Results and Discussion
5 Conclusions
References
Deep Learning Technology for Automatic Burned Area Extraction Using Satellite High Spatial Resolution Images
1 Introduction
2 Problem Statement
3 Literature Review
4 Materials and Methods
4.1 Satellite Data
4.2 Preprocessing
4.3 Pansharpening and Spectral Index Calculation
4.4 Deep Learning Model
5 Experiment
5.1 Visual Analysis
5.2 Quantitative Analysis
6 Results and Discussion
7 Conclusions
References
Classification Methods of Heterogeneous Data in Intellectual Systems of Medical and Social Monitoring
1 Introduction
2 Review of Literature
3 Problem Statement
4 Materials and Methods
4.1 Development of Models Presentation of WS & WL of Heterogeneous MSM Data
4.2 Development of a Method for Matching Class Markers and Clusters in the Teaching of the Kohonen Network with a Partial Involvement of the Teacher
4.3 Development of a Method for Classification of SS and CM Heterogeneous MSM Data
5 Conclusions and Future Work
References
IaaS-Application Development for Paralleled Remote Sensing Data Stream Processing
1 Introduction
2 Problem Statement
3 Literature Review
4 Algorithm Development
4.1 Data Loading and Batch Calculation
4.2 Kubernetes Cluster and Workload Calculation
4.3 Workflow Scheduling
5 Experiment, Results and Discussion
5.1 Neural Networks
5.2 Experimental Datasets
5.3 Simulation Results, Computation Cost and Discussion
6 Conclusions
References
Author Index