This book presents new perspective research results: models, methods, algorithms and applications in the field of Artificial Intelligence (AI). Particular emphasis is given to the medical applications - medical images recognition, development of the expert systems which could be interesting for the AI researchers as well for the physicians looking for the new ideas in medicine. The central audience of the book are researchers, industrial practitioners, students specialized in the Artificial Intelligence.
The Chapter "Development of a Program Code Review System Using Machine Learning Methods" presents a description of the developed approach and service for analyzing source code in Python. The service reduces the time for code review due to partial automation. The FastText algorithm is used to obtain vector representations of source code texts. A pre-trained neural network language model based on the transformer architecture was used to derive a possible natural language function assignment. A classifier based on the gradient boosting algorithm was used to detect duplicate PR. The developed service checks the changeset and publishes error and duplicate reports in changeset comment format after the changeset is published to a remote Git repository. The conducted testing did not reveal any errors that affect the operation. All the main functions of the system are performed correctly.
The Chapter "Production Control Based on a Quality Guarantor Computer Vision System" presents a solution for an intelligent quality control system capable of identification of product failures and manufacturing defects using a computer vision system. The proposed solution expands the capabilities of automated quality management systems in modern production by supporting operator actions using an artificial neural network. The resulting solution is called “intelligent quality guarantor”, being proposed and described in this paper it implements a new concept of production process multiple parameters monitoring and recognition using an intelligent computer vision system. Special software was developed capable of dynamic tracking of quality control process to decompose and classify the tasks being performed by patterns of deviance. The computer vision application being developed collects real-time data from cameras and analyzes data streams using machine learning algorithms. Based on predetermined quality standards, the system is capable of automatic sorting, as well as detecting visual defects corresponding to mechanical damage or manufacturing defects.
The Chapter "Deep Learning Approach to Recognition of Car Driver's Closed Eyes for Safety Reasons": Today it is difficult to overestimate the role of artificial intelligence systems in various sectors of the economy. One of the rapidly developing areas in the period of digitalization is the sphere of transport security. The development of unmanned vehicles is promising, but such systems raise many questions about the perpetrators in the event of various accidents. In this regard, at present, the vast majority of cars are driven by a person, which leads to corresponding problems. One of the key problems is the high proportion of accidents associated with falling asleep at the wheel. This article is aimed at solving the problem of analyzing the driver's condition using a video camera installed in the car. The system uses approaches from both traditional machine learning methods and deep learning methods based on convolutional neural networks. To detect eyes with a completeness close to 100%, in the absence of interfering objects, Haar cascades are proposed. For this, simple geometric calculations are performed based on the detected face, which significantly speeds up the process. In this case, a parallel system of two convolutional networks is used to recognize closed and open eyes. The completeness of recognition of closed eyes in one frame in the developed system is almost 90%.
Author(s): Olga Dolinina, Igor Bessmertny, Alexander Brovko, Vladik Kreinovich
Series: Studies in Systems, Decision and Control
Publisher: Springer
Year: 2023
Language: English
Pages: 694
Organization
Preface
Contents
The Study of Artificial Intelligence
Semantic Modeling of Predicate Calculus Based on N-Tuple Algebra
1 Introduction
2 Axioms and Inference Rules of Predicate Calculus
3 Interpretation of Language of Mathematical Logic
4 N-Tuple Algebra
5 Interpretation of Logical Inference
6 Unresolved Issues
7 Positive Results of Using NTA
8 Conclusion
References
Method of Iterative-Order Optimization of Multicriteria Problems Using the Local Importance of Criteria
1 Introduction
2 Basic Definition and Notation
3 Multicriteria Optimization Approaches Classification
4 Complexity of Questions (Operations) in Multicriteria Methods
5 An Iterative Optimization Approach
6 An Iterative Structured Optimization Approach
7 Development of an Iterative Approach in the Form of an Iterative-Order Structured Optimization Method
8 Algorithm for Iterative-Order Optimization of Multicriteria Problems Using Qualitative Global and Local Importance
9 Analysis and Comparison of Decision Support Systems
10 Verification of the Method Using a Computer System for Supporting the Adoption of Multi-criteria Decisions with Support for the Local Importance of Criteria
10.1 General Description
10.2 An Example Using Local Importance
10.3 An Example Without Using Local Importance
11 Conclusion
References
Neural Network Representation for Ordinary Differential Equations
1 Introduction
2 Problem Statement
3 PNN Design
3.1 General Architecture
3.2 PNN Initialization
4 Numerical Experiments
4.1 Forward Problem
4.2 Inverse Problem
5 Conclusion
References
Modeling Situations in Spatial Analysis
1 Introduction
2 Analysis of Known Approaches to Transforming Situations
3 Properties of Transformation Method
4 Implementation of Transformation Algorithms
5 Conclusion
References
Development of a Program Code Review System Using Machine Learning Methods
1 Introduction
2 Existing Source Code Analysis Approaches
3 Source Code Review Using Neural Networks
4 Software Service Architecture
5 Experimental Results
6 Conclusions
References
Production Control Based on a Quality Guarantor Computer Vision System
1 Introduction
2 State of the Art
3 Intelligent Quality Guarantor
4 Implementation Results
5 Conclusion
References
Assessment of the Vine–Copula for the Gamma Poisson Model of Risky Person's Behavior with Synthetic Data
1 Introduction
2 Background
2.1 The Last Episodes Methodology
2.2 Copulas and Vines
2.3 C–vine for the Gap Lengths in the Context of the Gamma Poisson Model of Behavior
3 Assessing the Types of Copula for the Gamma Poisson Model of Behavior: Synthetic Data and Numerical Experiments
4 Discussion and Result
5 Conclusion
References
Fuzzy Models and Soft Calculations
Implementation of the Constraints Principle in Problems of Multicriteria Choice Under Conditions of Fuzzy Assignment of Criteria Values
1 Introduction
2 Multicriteria Selection Method Based on the Constraints Principle
3 Modification of the Method with Fuzzy Criteria
4 Fuzzification of the Modified PSEM Problem
5 Calculation of Values of Gravity Centers for Triangular Membership Functions
6 Modelling Results
7 Conclusion
References
Divergence of Comparative Possibility Distributions as a Numerical Measure of How Difficult it is to Reach a Consensus
1 Introduction
2 Preliminary Notes: The Lattice of Comparative Possibility Distributions
3 Divergence of Comparative Possibility Distributions
3.1 Definition
3.2 Algorithm
3.3 Example
4 Application to Renewable Energy Project Evaluation
5 Discussion
5.1 Comparison with the Kendall Tau Distance [32]
5.2 Relationship with the Jaccard Index [33]
5.3 Comparison with Voting Algorithms
6 Conclusion
References
Data Mining, Knowledge Engineering and Ontology
Case-Based Reasoning and Ontology-Based Approach to Selecting Equipment Solutions in Oilfield Engineering
1 Introduction
2 Methods and Models
2.1 Case Based Reasoning
2.2 CBR in Selecting and Configuring Oil and Gas Facilities
2.3 An Ontological Model for Describing Cases and Suggesting Solutions
3 Application
4 Discussion
5 Conclusion
References
Associative Classification Based on the Table Constraint Satisfaction
1 Introduction
2 Table Constraint Satisfaction Problems
3 The Method of Solution of the Closed Patterns Discovery Problem
4 Frequent Pattern Discovery
5 Association Rule Mining
6 Conclusion
References
Extracting Entity Relationship from Arabic Texts
1 Introduction
2 Related Work
3 Main Work
3.1 Language Pre-processing
3.2 Tokenization
3.3 Named Entity Recognition
3.4 Assembling Entity’s Name
3.5 Extraction of Relations
4 Results
5 Conclusion
References
A Probabilistic-Statistical Approach to Detection of Semantic Relations Between Indexing Terms
1 Introduction
2 Related Works and Theoretical Analysis
3 Probabilistic-Statistical Model
4 Conclusions
References
Creating a Metagraph Dataset Based on the ATOMIC Dataset
1 Introduction
2 The Example of the Annotating Metagraph Model
3 Restrictions of the RDF Model
4 The ATOMIC Dataset and Its Transformation to RDF
5 Markov Clustering Algorithm
6 Breadth Clustering Algorithm
7 Experiments
8 Conclusions
References
TableProcessor: The Tool for the Analysis and the Interpretation of Web Tables to Create the Geo Knowledge Base of Kazakhstan
1 Introduction
2 Results and Discussion
3 Conclusion
References
Features of Knowledge Representation of Automatic Control Systems Design Methods
1 Introduction
2 Proposed Solution
3 Using of Knowledge in the Intelligent Systems for Supporting Design Decision Making of Intelligent Control Systems
4 Results
5 Conclusion
References
Computational Linguistics and the Semantic Web
Process-Semantic Analysis of Words and Texts
1 Introduction
2 Methods
2.1 Process-Semantic Template
2.2 Extraction of Process Semantics from Word2vec Data
3 Results
3.1 Individual Words
3.2 Texts
4 Discussion
References
Topic Modeling Algorithm Using Quantum Probability Theory
1 Introduction
2 Methods
2.1 Probabilistic Topic Modeling Using Methods of Quantum Mathematics
2.2 Implementation of Quantum Probabilistic Latent Semantic Analysis
3 Results
4 Conclusion
References
Image and Sound Processing, Image Recognition and Computer Vision
Method for Detecting Person in a Frame of Video Stream and Classifying his Position Using an Artificial Neural Network
1 Introduction
2 Method
2.1 Formulation of the Problem
2.2 Discovery Stage
2.3 Classification Stage
2.4 Method Implementation
3 Evaluation of Network
4 Conclusion
References
Deep Learning Approach to Recognition of Car Driver's Closed Eyes for Safety Reasons
1 Introduction
2 Brief Overview of Transport Monitoring Systems
3 Materials and Methods
4 Modeling Convolutional Neural Networks
5 Working with Video Stream
6 Conclusions
References
Russian Sign Language Recognition Using MediaPipe
1 Introduction
2 Sign Language Recognition
2.1 Research Methodology
2.2 Problem Statement
2.3 Russian Sign Language
2.4 Video Dataset for Russian Sign Language
2.5 MediaPipe Detection
2.6 Dynamic Time Warping Algorithm
2.7 Implementation of Models
2.8 Sign Prediction
2.9 Results and Analysis
3 Conclusion/Future Work
References
Deep Machine Learning Techniques in the Problem of Estimating the Expression of Psychological Characteristics of a Social Media User
1 Introduction
2 Problem Statement
3 Data and Methods
4 Results of Facial Recognition Methods
5 Transfer Learning and Assessment of Psychological Defenses
6 Discussion and Results
7 Conclusion
References
Development of Software for Managing Treadmills Based on Computer Vision
1 Introduction
2 Methods
2.1 System Analysis of Objects in the Subject Area
2.2 Statement of the Research Problem
2.3 An Algorithm for Solving the Control Problem Based on Computer Vision
3 Results
3.1 Implementation of Treadmill Management Software
3.2 Development of a Neural Network for Recognizing Human States
3.3 Experimental Studies on the Application of the Developed Software
4 Discussion
5 Conclusion
References
3D Artifact Localization Using Connected Components
1 Introduction
2 Previous Work
3 Methodology
3.1 Dataset
3.2 Artifact Segmentation Algorithm
4 Results and Discussion
5 Conclusion
References
Applied Intelligent Systems
The Concept of Intelligent Support for Laser Additive Manufacturing Process Engineer
1 Introduction
2 Case Study
3 Research Approach and Results
3.1 Information Resources of the LBAM Knowledge Portal
3.2 Software Components of the LBAM Knowledge Portal
4 Conclusion
References
Socio-Economic Sustainability Monitoring Based on Intelligent Analysis of Social Media
1 Introduction
2 State of the Art
3 Social-Economic Sustainability Model
4 Analysis of the Mutual Influence of Factors and Hidden Patterns
5 Implementation in Healthcare
6 Conclusion
References
Predicting the Optimal Operation of Burners Based on Random Forest
1 Introduction
2 Development of the Digital Twin
2.1 Platform Concept
2.2 Development of Solid Models
3 Results
4 Conclusions
References
Approach to Building an Intelligent Repository of Software Projects
1 Introduction
2 Model of Representation of a Software System Project Repository as an Element of a Search Index
3 Software Projects Indexing Method
3.1 Algorithm for Extracting Entities and Business Processes from a Source Code as Context Features
3.2 Algorithm for Extracting Development Process Metrics
4 Method for Evaluating the Similarity of a Repository and a Search Query
5 Conclusion
References
Models and Methods for Searching Cohesive Teams of Software Developers
1 The Analysis of IT Personnel Market
2 The Model of Cohesive Team and the Overview of Related Methods
3 The Architecture of Implemented Software
4 Tests
5 Summary
References
Anomaly Detection in a Granulation Plant for Uranium Dioxide Powder Pressing into Fuel Pellets
1 Introduction
2 Technological Process
3 Data
3.1 Raw Data
3.2 Preprocessing of Raw Data
4 Methods
4.1 Model
4.2 Detection Algorithm
4.3 Upper Control Limit Selection
5 Results and Conclusion
References
Artificial Intelligence Systems in the Field of Safety at Work and Life
1 Introduction
2 On the Methods of Ensuring a High-Quality Transitional Process in Control
3 Analysis
4 Confirmation of the Above Reasoning by Modeling
5 Solving the Same Problem by Numerical Optimization
6 Discussion and Conclusion
References
Ensuring the Robustness of Modern Mechatronic Systems Using Artificial Intelligence Methods
1 Introduction
2 Functional/Kinematic Diagram of a Digitally Controlled Test Bench with Angular Velocity and Linear Acceleration Meters
3 Synthesis of Controllers for Inertial Sensing Elements in a Precision Test Bench Using Artificial Intelligence Methods
4 Controller Synthesis for a Precision Test Bench with ISE Using the Terminal Control Methods Based on an Observer in the Form of Artificial Neural Network
5 Terminal Control of the Test Bench Platform Based on a Hybrid Neural Controller
6 Conclusions
References
Service-Oriented Architecture for Block-Base Construction Building
1 Introduction
2 Distributed Computing Concept on Prolog
3 Description of Interserver Communication
4 Results
5 Conclusion
References
The Use of Entropy and Divergence of Kullback–Leibler in Systems of Intellectual Analysis of Test Results
1 Introduction
2 Survey Methodology
3 Results and Discussion
4 Conclusion
References
Artificial Intelligence in Industry, Energy and Radio Engineering
Preventive Analytics in Safety Tasks for Critical Infrastructures Development
1 Introduction
2 Main Features of the SMS
2.1 Situation Processing
2.2 Metrification of the State Space
2.3 Scenarios in the Generalized State Space
3 Proactive Safety Analytics for INCs
3.1 Digital Twins for INCs
3.2 Situational Awareness in INCs
3.3 Situational Awareness in Net-Centric Systems
4 Approbation of the SMS
5 Conclusion
References
Mathematical Model for Training a Neural Network Used to Predict Atmospheric Pollutants from Vehicles
1 Introduction
2 Problem
3 Computational Experiment
4 The Mechanism of Using the Neural Network in Decision-Making Process
5 Conclusion
References
Rule-Based Model for Describing the Processes of Rise and Transfer of Anthropogenic Emissions in the Atmosphere
1 Introduction
2 Problem
3 Mathematical Model
4 Computational Experiments
5 Practical Implementation of the Developed Software
6 SOTA Evaluation of the Solution of the Problem
7 Conclusion
References
Using Convolutional Neural Networks for Training Forest Machine Operators
1 Introduction
2 Related Work
3 Materials and Methods
4 The Algorithm for Changing of the Dynamic Parameters in Virtual Reality
5 The Results and Discussion
6 Conclusions
References
Artificial Intelligence in Medicine and Bioengineering
Computer Analysis of Lumbar Spine Magnetic Resonance Imaging Scans via Neural Network Algorithms
1 Introduction
2 Review of Actual Researches
3 Development of the Segmenting Neural Network
4 Description of Quantitative Parameters of Intervertebral Discs
5 Development of the Neural Network for the Classification of the Pathology of the Intervertebral Discs
6 Aggregation of the Described Methods and Models for Processing MRI Scans
7 Conclusion
References
Using of Artificial Intelligence in Detection of the Compliance of the Ophthalmic Patients
1 Introduction
2 Specifics of Implementation of the Expert System for Compliance Prediction
3 Structure of the ES “Glaucoma Compliant”
4 Results of Testing and Implementation of ES “Glaucoma Compliant”
5 Conclusion
References
An Overview of Using Deep Learning Algorithms for Anemia Detection
1 Introduction
2 Deep Learning Approaches
2.1 Eye Conjunctiva Color as Anemia Sign
2.2 Retinal Fundus Images and Anemia
2.3 Detect Anemia with ECGs
3 Smartphone Apps for Anemia Detection
3.1 Fingernail Color and Anemia
3.2 Photoplethysmography
3.3 HemoSmart Anemia Detection
4 Methods to Improve the Accuracy of Deep Learning Algorithms in the Detection of Anemia
4.1 Ridge Regression Approach to Detect Anemia
4.2 Image Preprocessing Filtering
4.3 Semantic Segmentation of the Conjunctiva Region as a Preprocessing Method
5 Conclusion
References
Expert System KORDEX for Solving of Urgent Cardiology Problems
1 Introduction
2 Methodology for Solving the Problem of Predicting an Unfavorable Outcome of Unstable Angina
3 Debugging of the Expert System KORDEX-2
4 Conclusion and Discussion
References
Application of Convolutional Neural Networks in the Analysis of Microimages of Skin Neoplasms in the Early Diagnosis of Melanoma
1 Introduction
1.1 The Early Melanoma Diagnosis Task
1.2 Application of Information Technologies and Artificial Intelligence Technologies in the Diagnosis of Melanoma
2 Using a Convolutional Neural Network to Find Objects in An Image
2.1 Description of the Neural Network Model Architecture
2.2 Data Preprocessing Before Model Training
2.3 Neural Network Learning Process
3 Experimental Results
3.1 Segmentation Module Experiments
3.2 Analysis Module Experiments
4 Conclusion
References
The Possibilities of Artificial Intelligence to Control the Adequacy of Hypovolemia Correction
1 Introduction
2 Methods
3 Results
3.1 Theoretical Research
3.2 Experimental Studies
3.3 Clinical Trials
4 Conclusion
References
Automated ABCDE Image Analysis of a Skin Neoplasm with Interpretable Results
1 Introduction
1.1 The Early Melanoma Diagnosis Task
1.2 Application of Information Technologies in the Early Melanoma Diagnosis Problem
1.3 Neural Network Approaches to Automating ABCDE Analysis
2 The Proposed Approach for Calculating the Metrics of ABCD Analysis
2.1 Image Preprocessing
2.2 Metric Value Calculation
2.3 Calculation of Threshold Coefficients
3 Experiments on the Calculation of Threshold Coefficients
4 Conclusion
References
Smart City Technologies and Internet of Things
Timeliness of Multipath Redundant Transmissions When All Paths are not Accessible for Some Request Sources
1 Introduction
2 Network Structure with Duplicate Paths
3 Conclusion
References
Off-Road Autonomous Vehicle Control System
1 Literature Review
2 Architectural Diagram of the Control System
3 Data Processing Function Blocks
3.1 Localization
3.2 Semantic Segmentation
3.3 Obstacle Detection
3.4 Mapping Service
3.5 Path Planning
3.6 Car Control
4 Prototype of the Car Control System
5 Conclusion
References
Adaptive City Traffic Lights As Collective IoT Devices
1 Introduction
2 Overview of Methods for Improving Urban Traffic
3 Microcontroller Method for Improving Smart Intersections
4 Optimization of Throughput Capacity of Highways in Almaty with the Help of Collective IoT Traffic Lights on the Example of the Dzhandosov/Zharokov Intersection
5 Conclusion
References