Rough sets : international joint conference, IJCRS 2021, Bratislava, Slovakia, September 19-24, 2021 : proceedings

Preface
Organization
Contents
Invited Papers
Mining Incomplete Data Using Global and Saturated Probabilistic Approximations Based on Characteristic Sets and Maximal Consistent Blocks
	1 Introduction
	2 Incomplete Data
	3 Probabilistic Approximations
		3.1 Global Probabilistic Approximations Based on Characteristic Sets
		3.2 Saturated Probabilistic Approximations Based on Characteristic Sets
		3.3 Global Probabilistic Approximations Based on Maximal Consistent Blocks
		3.4 Saturated Probabilistic Approximations Based on Maximal Consistent Blocks
		3.5 Rule Induction
	4 Experiments
	5 Conclusions
	References
Determining Tanimoto Similarity Neighborhoods of Real-Valued Vectors by Means of the Triangle Inequality and Bounds on Lengths
	1 Introduction
	2 Basic Notions and Properties
		2.1 The Euclidean Distance, the Cosine Similarity and the Tanimoto Similarity
		2.2 ε-Neighborhoods and k Nearest Neighbors
	3 Using the Triangle Inequality Property to Calculate Euclidean and Cosine ε-Neighborhoods
		3.1 Using the Triangle Inequality to Calculate Euclidean ε-Neighborhoods
		3.2 Calculating Cosine ε-Neighborhoods by Means of the Triangle Inequality
	4 Using Bounds on Vector Lengths to Calculate Tanimoto Similarity ε-Neighborhoods
	5 Calculating Tanimoto ε-Neighborhoods by Means of the Triangle Inequality
	6 Calculating Tanimoto ε-Neighborhoods by Means of the Triangle Inequality and Lengths of Vectors
	7 Summary
	References
Rough-Fuzzy Segmentation of Brain MR Volumes: Applications in Tumor Detection and Malignancy Assessment
	1 Introduction
	2 Segmentation of Brain MR Images
	3 Brain Tumor Detection and Gradation
	References
DDAE-GAN: Seismic Data Denoising by Integrating Autoencoder and Generative Adversarial Network
	1 Introduction
	2 Related Work
		2.1 Seismic Noise Reduction Methods
		2.2 Noise Modeling Based Denoising Methods
		2.3 AutoEncoder Based Denoising Methods
	3 DDAE-GAN Based Blind Denoiser
		3.1 Paried Data Constructing
		3.2 Pre-training
		3.3 Transfer Learning
	4 Examples
		4.1 Synthetic Examples
		4.2 Field Examples
	5 Conclusions
	References
Classification of Multi-class Imbalanced Data: Data Difficulty Factors and Selected Methods for Improving Classifiers
	1 Introduction
	2 Related Works on Classification of Multi-class Imbalanced Data
	3 Difficulty Factors in Imbalanced Data
		3.1 Earlier Studies on Binary Imbalanced Classes
		3.2 Multi-class Difficulties
	4 Identifying Types of Examples in Multi-class Imbalanced Data
	5 Discovering Split of Classes into Sub-concepts and Rare Examples
	6 Multi-class Hybrid Resampling Algorithm SOUP
	7 Multi-class Variant of BRACID Algorithm
		7.1 Rule Induction from Binary Imbalanced Data with BRACID
		7.2 Generalizing BRACID for Multiple Imbalanced Classes
	8 Multi-class Extension of Bagging Ensemble
	9 Software Implementations of Specialized Algorithms for Multi-class Imbalanced Data
	10 Future Research Directions and Conclusions
	References
Core Rough Set Models and Methods
General Rough Modeling of Cluster Analysis
	1 Introduction
		1.1 Background
	2 New Rough Semantic Approaches
	References
Possible Coverings in Incomplete Information Tables with Similarity of Values
	1 Introduction
	2 Rough Sets from Coverings in Complete Information Tables
	3 Rough Sets from Possible Coverings in Incomplete Information Tables
	4 Conclusions
	References
Attribute Reduction Using Functional Dependency Relations in Rough Set Theory
	1 Introduction
	2 Preliminaries
		2.1 Rough Sets
		2.2 Reducts for Information Systems
		2.3 Functional Dependency Relations
		2.4 Closure Operators
		2.5 Relationships on Attribute Sets
	3 Functional Dependency Relations
	4 Conclusions
	References
The RSDS: A Current State and Future Plans
	1 Generally About the RSDS
	2 Further Plans
	3 Final Remarks
	References
Many-Valued Dynamic Object-Oriented Inheritance and Approximations
	1 Introduction
	2 Preliminaries
		2.1 Many-Valued Logics
		2.2 Nested Structures
		2.3 Rule-Based Object-Oriented Query Languages
	3 Many-Valued Dynamic Object Inheritance
	4 Approximations
	5 Conclusions
	References
Related Methods and Hybridization
Minimizing Depth of Decision Trees with Hypotheses
	1 Introduction
	2 Decision Tables
	3 Decision Trees
	4 Construction of Directed Acyclic Graph (T)
	5 Minimizing the Depth of Decision Trees
	6 Results of Experiments
	7 Conclusions
	References
The Influence of Fuzzy Expectations on Triples of Triangular Norms in the Weighted Fuzzy Petri Net for the Subject Area of Passenger Transport Logistics
	1 Introduction
	2 Fuzzy Expectations for wFPN
	3 The Review of wFPN Model for the Experiment on Triples of Functions
	4 The Influence of Fuzzy Expectations on the Results of the wFPN Model
	5 Conclusions
	References
Possibility Distributions Generated by Intuitionistic L-Fuzzy Sets
	1 Introduction
	2 Preliminaries
		2.1 Algebraic Structures of Truth Values
		2.2 Intuitionistic Fuzzy Sets and Intuitionistic L-fuzzy Sets
	3 From Intuitionistic L-Fuzzy Sets to Possibility Distributions
		3.1 Possibility Distributions
		3.2 Possibility Distributions Generated by Intuitionistic L-fuzzy Sets
		3.3 Possibility Distributions Generated by Intuitionistic L-fuzzy Sets Based on an IMTL-algebra
	4 From Possibility Distributions to Intuitionistic Fuzzy Sets
		4.1 An Algorithm to Find the Intuitionistic L-fuzzy Set Generating a Given Possibility Distribution
	5 Conclusions and Future Directions
	References
Feature Selection and Disambiguation in Learning from Fuzzy Labels Using Rough Sets
	1 Introduction
	2 Background
		2.1 Possibility Theory
		2.2 Rough Set Theory
		2.3 Belief Function Theory
	3 Possibilistic Decision Tables and Reducts
		3.1 Possibilistic Decision Tables
		3.2 Possibilistic Reducts
		3.3 Entropy Reducts
	4 Conclusion
	References
Right Adjoint Algebras Versus Operator Left Residuated Posets
	1 Introduction
	2 Preliminaries
		2.1 Dedekind-MacNeille Completion
		2.2 Algebraic Structures
	3 Adjoint Property in Operator Left Residuated Posets
		3.1 Extension of M and R to 2P
		3.2 Requirements for a Proper Fuzzy Modus Ponens
		3.3 Extension of Operator Left Residuated Posets
	4 Operator Left Residuated Posets from a Dedekind-MacNeille Completion
	5 Conclusions and Future Work
	References
Adapting Fuzzy Rough Setspg for Classification with Missing Values
	1 Introduction
	2 Interval-Valued Fuzzy Rough Sets
	3 FRNN with Interval-Valued Approximations
	4 Conclusion
	References
Areas of Applications
Spark Accelerated Implementation of Parallel Attribute Reduction from Incomplete Data
	1 Introduction
	2 Preliminaries
		2.1 Apache Spark Computing Model
	3 Spark Parallelization of Attribute Reduction from Incomplete Data
	4 Experimental Evaluation
		4.1 Selection of the Number of Data Partitions
		4.2 Evaluation of the Parallelism Metrics
	5 Conclusions
	References
Attention Enhanced Hierarchical Feature Representation for Three-Way Decision Boundary Processing
	1 Introduction
	2 Proposed Method
	3 Performance Evaluation
	4 Conclusion
	References
An Opinion Summarization-Evaluation System Based on Pre-trained Models
	1 Introduction
	2 Related Works
	3 An Opinion Summarization-Evaluation Algorithm
		3.1 Subjective Analysis and Opinion Mining
		3.2 Hierarchical Metrics
	4 Experiments and Analysis
		4.1 Experimental Settings
		4.2 Experiment Results and Analysis
	5 Conclusion and Future Works
	References
Fuzzy-Rough Nearest Neighbour Approaches for Emotion Detection in Tweets
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Data Cleaning
		3.2 Tweet Embedding
		3.3 Similarity Relation
		3.4 Classification Methods
		3.5 Evaluation Method
	4 Experiments
		4.1 Detecting the Best Setup for Embeddings
		4.2 Ensembles
	5 Results on the Test Data
	6 Conclusion and Future Work
	References
Three-Way Decisions Based RNN Models for Sentiment Classification
	1 Introduction
	2 Related Work
		2.1 RNN Models
		2.2 Three-Way Decisions
	3 The Proposed Method
		3.1 Algorithm
		3.2 Probability Adjustment Strategies
	4 Experiment
		4.1 Datasets and Baseline Methods
		4.2 Experimental Results
		4.3 Parameter Analysis
	5 Conclusion
	References
Tolerance-Based Short Text Sentiment Classifier
	1 Introduction
	2 Data Sets
	3 Models
		3.1 Tolerance Near Sets
		3.2 Transformer Model
	4 Experiments and Analysis of Results
	5 Conclusion
	References
Knowledge Graph Representation Learning for Link Prediction with Three-Way Decisions
	1 Introduction
	2 Related Work
		2.1 Knowledge Graph Embedding Models
		2.2 Three-Way Decisions
	3 Our Approach
		3.1 Relation Neighbor
		3.2 Knowledge Representation with Three-Way Decisions
		3.3 Loss Function
	4 Experiments
		4.1 Datasets
		4.2 Baselines and Experiment Setting
		4.3 Evaluation Metrics
		4.4 Experiment Results
	5 Conclusion
	References
PNeS in Modelling, Control and Analysis of Concurrent Systems
	1 Introduction
	2 Preliminaries
		2.1 Petri Net
		2.2 The Ways of Working of a Petri Net
		2.3 Extensions of Petri Nets
		2.4 Hierarchical Petri Nets
	3 Properties of Petri Nets
		3.1 Behavioural Properties
		3.2 Structural Properties
	4 Analysis Methods
	5 PNeS
		5.1 General
		5.2 Behaviour Analyzers
		5.3 Controlling Robots
	6 Conclusion and Further Work
	References
3RD: A Multi-criteria Decision-Making Method Based on Three-Way Rankings
	1 Introduction
	2 Three-Way Ranking Based Multi-criteria Decision-Making
		2.1 Trisecting the Set of Decision Alternatives
		2.2 Computing Dominance Values of Decision Alternatives
		2.3 3RD Method
	3 An Illustrative Example
		3.1 An Example Solved by the TODIM method
		3.2 The Procedure of 3RD Method
		3.3 Sensitivity Analysis of 3RD
	4 Conclusion
	References
Author Index