Artificial Intelligence: 19th Russian Conference, RCAI 2021, Taganrog, Russia, October 11–16, 2021, Proceedings (Lecture Notes in Computer Science, 12948)

Description

This book constitutes the proceedings of the 19th Russian Conference on Artificial Intelligence, RCAI 2021, held in Moscow, Russia, in October 2021. The 19 full papers and 7 short papers presented in this volume were carefully reviewed and selected from 80 submissions. The conference deals with a wide range of topics, categorized into the following topical headings: cognitive research; data mining, machine learning, classification; knowledge engineering; multi-agent systems and robotics; natural language processing; fuzzy models and soft computer; intelligent systems; and tools for designing intelligent systems. Preface Organization Contents Cognitive Research Heterogeneous Formal Neurons and Modeling of Multi-transmitter Neural Ensembles 1 Introduction 1.1 Motivation 1.2 Contribution 2 Related Work 3 Model Formulation 3.1 Neurons and Their Interactions 3.2 Classes of Neural Activity 3.3 Network Dynamics 4 Biological Rhythms 4.1 Oscillation by Mutual Inhibition 4.2 Three-Phase Feeding Rhythm of a Pond Snail 5 Conclusion References Methods for Recognition of Frustration-Derived Reactions on Social Media 1 Introduction 2 Rosenzweig’s Frustration Test 3 Related Work 4 Pikabu Frustration Dataset 5 Features for the Frustration Detection 6 Methods and Results of the Frustration Detection 7 Conclusion References Identification of the Network State Based on the ART-2 Neural Network with a Hierarchical Memory Structure in Parallel Mode 1 Introduction 2 Description of the Structure of the ART-2m Network 3 Algorithm of ART-2m Network Functioning in Parallel Mode 4 Description of the Sample for Training 5 Experimental Results 6 Conclusion References Data Mining, Machine Learning, Classification Ranking Weibull Survival Model: Boosting the Concordance Index of the Weibull Time-to-Event Prediction Model with Ranking Losses 1 Introduction 2 Related Work 2.1 Concordance Index 2.2 Deep Learning in Survival Analysis 3 The Proposed Model 3.1 Modeling Time-to-Event with Weibull Distribution 3.2 General Framework 3.3 Improving Concordance by Adding Cross-Entropy Loss 3.4 Improving Concordance by Adding Contrastive Loss 4 Experiments 4.1 Data 4.2 Results 5 Conclusion References Predicting Different Health and Lifestyle Behaviors of Social Media Users 1 Introduction 2 Related Work 3 Dataset 4 Features 5 Method and Evaluation Setup 6 Results 7 Discussion 8 Conclusion References Methods for Finding Consequences with Specified Properties 1 Introduction 2 Briefs from N-tuple Algebra 3 Properties of “Interesting” Consequences 4 Methods for Calculating “Interesting” Consequences 5 Conclusion References Data Mining Methods for Analysis and Forecast of an Emerging Technology Trend: A Systematic Mapping Study from SCOPUS Papers 1 Introduction 2 Existent Viewpoints for Determining Promising Technologies 3 Systematic Mapping Study 3.1 Defining Research Questions (RQs) 3.2 Conducting Search 3.3 Refining Documents 3.4 Extracting Keywords from the Full Text 3.5 Extracting Data and Creating a Systematic Map 4 Basic Ideas of Patent Clustering Methods in TTAF Research 5 Conclusion References Machine Learning for Assessment of Cardiometabolic Risk Factors Predictive Potential and Prediction of Obstructive Coronary Arteries Lesions 1 Introduction 2 Materials and Methods 2.1 Patient Characteristics 2.2 Data Processing 2.3 Study Design 3 Results 4 Discussion 5 Conclusion References Knowledge Engineering Application of FCA for Domain Model Theory Investigation 1 Introduction 2 Methods of Formalization of Subject Domains 2.1 Problems of Domain Modeling and Related Works 2.2 Our Approach to Formalizing Subject Domains 3 Axiomatization of Domain Cases 4 Axiomatizable Classes of Fragments of Atomic Diagrams 5 Boolean-Valued Models Generated by Domain Case Classes 5.1 Weak Boolean-Valued Models 5.2 Using the Proposed Model-Theoretical Methods for Generating Knowledge About Subject Domains 6 Conclusion References The Metagraph Model for Complex Networks: Definition, Calculus, and Granulation Issues 1 Introduction 2 The Basic Provisions and Definitions of Information Granulation 3 The Metagraph Data Model 4 Using Metagraph Agents to Process Metagraph Model 5 The Metavertex as a Basic Structure of a Metagraph Model 5.1 Metagraph Vertex Representation 5.2 Metagraph Edge Representation 5.3 Metagraph Attribute Representation 5.4 Metagraph Fragment Representation 5.5 Metagraph Function Agent Representation 5.6 Metagraph Rule Agent Representation 5.7 Active Metagraph Representation 6 The Metagraph Calculus 6.1 Construction Operator 6.2 Deconstruction Operator 6.3 Deletion and Transitive Deletion Operators 6.4 Metagraph Calculus from the Point of View of the Atomicity Definition 7 The Granularity of the Metagraph Model 7.1 Atomicity Property 7.2 Holonic Organization 7.3 Construction and Deconstruction Operations for Granules 7.4 Higher-Order Information Granules 7.5 Interrelationship and Intrarelationship 7.6 Refinement and Coarsening and Partial Ordering 7.7 Similarity Relationship 7.8 Granulations as Partitions and Coverings 7.9 Measures of Granularity 7.10 Granularity of Metagraph Agents 8 Conclusions References Subjective Expert Evaluations in the Model-Theoretic Representation of Object Domain Knowledge 1 Introduction 2 Precedent Models as Formalization of Frequency Interpretation of Probability 3 Fuzzy Models as a Formalization of Subjective Probability 4 Expert's Subjective Knowledge 5 Algorithm for Checking the Correctness of Expert Knowledge 6 Conclusion References Multiagent Systems and Robotics Q-Mixing Network for Multi-agent Pathfinding in Partially Observable Grid Environments 1 Introduction 2 Problem Statement 3 Method 4 Experimental Environment for Multi-Agent Pathfinding 5 Experiments 6 Conclusion References Subdefinite Computations for Reducing the Search Space in Mobile Robot Localization Task 1 Introduction and Problem Statement 2 Probabilistic Approaches to the Localization Problem 3 Proposed Approach 3.1 Subdefinite Computations 3.2 Application to the Problem of Localization 3.3 Features of the Application 3.4 Localization Only by Bearing 3.5 Combination with Histogram Filter 4 Experiments 4.1 Simple Simulation 4.2 Gazebo Simulation 5 Conclusion References Enhancing Exploration Algorithms for Navigation with Visual SLAM 1 Introduction 2 Related Work 3 Problem Statement 4 Method Overview 4.1 SLAM 4.2 Exploration: Base Version 4.3 Exploration: Our Enhancements 4.4 Path Planner 4.5 Path Follower 5 Experiments 5.1 Experimental Setup 5.2 Results with RGB-D Input 5.3 Results with FCNN-predicted Depths 6 Conclusion and Future Work References Natural Language Processing Relying on Discourse Trees to Extract Medical Ontologies from Text 1 Introduction 2 Related Work 2.1 Ontology Extraction in Medical Domain 2.2 Discourse Organization of the Text 3 System Architecture 4 Candidate Ontology Entry Extractor 4.1 Discourse-Level Support for Ontology Construction 4.2 Rhetorical Relations Determining Informative Text Spans 4.3 Relation Extractor Based on Syntactic Parsing 5 Phrase Aggregator 6 Neural Dictionary Manager 7 Validating Ontology 8 Evaluation 8.1 Dataset 8.2 Assessment of Ontology Consistency 8.3 Assessment of Search Improvement Due to Ontology 9 Conclusion References TITANIS: A Tool for Intelligent Text Analysis in Social Media 1 Introduction 2 Related Work 3 TITANIS Design and Components 4 TITANIS Functionality 4.1 Psycholinguistic Analysis 4.2 Analysis of Emotional Structure 4.3 Analysis of Discourse Structure 4.4 Frustration Detection 5 Empirical Research 5.1 The Task 5.2 Pikabu.ru Dataset 5.3 Psycholinguistic Aspect 5.4 Semantic Aspect 5.5 Discourse Aspect 5.6 Frustration Level 6 Conclusion References Approach to the Automated Development of Scientific Subject Domain Ontologies Based on Heterogeneous Ontology Design Patterns 1 Introduction 2 The ODPs Ontology 3 The Architecture of the System for the Automated Development of SSD Ontologies 4 Automatic Generation of Lexico-Syntactic Patterns 4.1 Lexico-Syntactic Patterns 4.2 Principles of LSP Generation 4.3 LSP Generation Algorithm 5 LSP-Based Ontology Population 6 Conclusion References Fuzzy Models and Soft Computing The PC-Algorithm of the Algebraic Bayesian Network Secondary Structure Training 1 Introduction 2 Related Works 3 PC-Algorithm 4 Generation of an Algebraic Bayesian Network Based on the Bayesian Belief Network 5 PC-Algorithm for Algebraic Bayesian Networks 6 Conclusion References Logistic-Based Design of Fuzzy Interpretable Classifiers 1 Introduction 2 Elements of DS Theory 3 Classification Model Based on Logistic Regression 4 Evidential Classifier 5 Linguistic Scale for Hypotheses of Class Membership 6 Combination of Simple m-Functions 7 Calibration of the Linguistic Scale 8 Linguistic Log-Regression Model 9 LLRds-Based Inference 10 Conclusion References Intelligent Systems Knowledge-Based Diagnostic System With a Precedent Library 1 Introduction 2 Overview of Diagnostic Systems for Hereditary Diseases 3 Materials and Methods 3.1 Precedents Representation Methods 3.2 Synthetic Precedents Constructing 3.3 Synthetic Precedents in Diagnostics as an Addition to the Expert Knowledge Base 4 Result: The Hybrid Service 4.1 Examples of Using the Hybrid Approach 5 Conclusion References Semiotic Models in Monitoring and Decision Support Systems 1 Introduction 2 Semiotic Model of the Situation 3 Methods and Approaches for Natural Language Processing 4 The Architecture of a Semiotic Decision Support System 4.1 Situation Monitoring 4.2 Decision-Making Based on a Semiotic Model of the Situation 5 Experiments 6 Conclusion References Cognitive Patterns for Semantic Presentation of Natural-Language Descriptions of Well-Formalizable Problems 1 Introduction 2 General Description 3 Materials and Methods 4 Results and Discussion 5 Related Studies 6 Conclusion References Detecting Anomalous Behavior of Users of Data Centers Based on the Application of Artificial Neural Networks 1 Introduction 2 Existing Methods of Analysis and Search for Anomalies 3 Obtaining Initial Data for an Artificial Neural Network 4 The Structure of the Analytical Unit for Detecting Anomalies 5 Detection of Anomalies in the Actions of DC Users 6 Conclusion References Tools for Designing Intelligent Systems A Study of the Feasibility of Creating of a Real-Time Neural Network Infrared Ground Objects Recognition System 1 Introduction 2 Research Procedure 3 Research Results 4 Conclusion References The Implementation of the Ontological Approach to Control of the Processes of Designing Integrated Expert Systems Based on the Problem-Oriented Methodology 1 Introduction 2 Some Features of Intellectualization of IES Prototyping Processes 3 Description of the Basic and Modified Ontology Models 4 Features of the Construction of an Applied Ontology of Typical IES Architectures 5 Conclusion References A Module for Industrial Safety Inspection Planning Based on Self-organization 1 Introduction 2 The Architecture of the Multicomponent Information System 3 Ontology 4 Scheduler 5 Local Rules 6 Implementation Example 7 Conclusion References Author Index