Information techniques in systemic research

This book is devoted to research and practical problems in the field of IT support decision-making processes and control in complex technical, economic-social and biological systems. Modern advances in computer technology have led in this area not only to the development of methods that streamline classical procedures, but also the development of innovative specialist methods in which numerical techniques play a decisive role, and methods that could only be applied thanks to widespread computerization, automation of measurement processes and globalization of information systems connected on the network.
The range of topics considered in the systemic studies is very wide. Describing them all is simply impossible. A solution in this situation is the gathering of works presenting these research areas, in which Polish scientists with a recognized international reputation are significant. Many of these areas are so far absent from Polish-language literature. This book can thus serve as a modern academic textbook supporting the didactic process at the higher years of master's studies and doctoral studies of technical, economic and university faculties. It can also be a source of inspiration for cognitive and research activities.

Table of Contents

Part I. System modeling

1. System modeling as a way to organize knowledge

Andrzej P. Wierzbicki

1.1. Admission
1.2. The role of mathematical modeling and computing in science in the twentieth century
1.3. Two basic concepts: chaos and complexity
1.4. Civilization of information and knowledge: basic trends
1.5. A change in the perception of the world
1.6. Various concepts of knowledge
1.7. Knowledge and modeling
1.8. A rational theory of intuition
1.8.1. Ontological and epistemological consequences of the rational theory of intuition
1.8.2. Strategic intuitive processes and creation of knowledge
1.9. The necessity to create new approaches in epistemology
1.10. Conclusions
Literature

2. Uncertain variables and their applications in uncertain systems

Zdzisław Bubnicki

2.1. Admission
2.2. Logic and uncertain variables
2.3. The problem of analysis
2.4. A parametric decision problem
2.5. Non-parametric decision problem
2.6. Generalization: soft variables
2.7. Stability of the dynamic system with uncertain parameters
2.8. Allocation problems and project management
2.9. System with uncertain and random parameters
2.10. Other problems and applications
2.11. Final remarks
Literature

3. Methodology for knowledge-based development management based on risk

Roman Kulikowski

3.1. Admission
3.2. The usefulness of sustainable development as a measure of development options
3.3. The usefulness of traditional and innovative technologies
3.4. Supporting management with regard to operational risk
3.5. Supporting the development with financial leverage, taking into account credit risk
3.6. Development perspectives of UTR methodology and applications
Literature

4. Nuclear estimators in the field of systemic research

Piotr Kulczycki

4.1. Admission
4.2. Methodology of constructing nuclear estimators
4.2.1. Selection of the kernel form
4.2.2. Specify the value of the smoothing parameter
4.2.3. Additional procedures
4.2.4. Modification of the smoothing parameter
4.2.5. Linear transformation
4.2.6. Restriction of the medium
4.2.7. Binary coordinates
4.2.8. The number of samples
4.2.9. Comments and comments
4.3. Sample applications for system tests
4.3.1. Elements of decision theory: Bayesian and minimax rules
4.3.2. Recognition of atypical elements
4.3.3. Sharpening of non-precision information
4.3.4. Parametric identification
4.3.5. Determining the spatial distribution of demand
4.4. Summary
Literature

Part II. Information analysis and processing

5. Statistical learning systems

Jacek Koronacki

5.1. Introductory remarks
5.2. Regression analysis
5.2.1. Global parametric models - from linear models to generalized models
linear and non-linear
5.2.2. From local smoothing to adaptive modeling
5.3. Classification under supervision
5.3.1. Linear methods and their generalizations
5.3.2. Bayesian classification and based on NW method, the nearest neighbors method,
classification trees
5.4. Threads and areas omitted
5.5. Instead of a conclusion
Literature

6. Fuzzy-neural methods for analysis and processing

data

Danuta Rutkowska

6.1. Introduction
6.2. Data grouping and classification
6.3. Fuzzy-neural grouping and classification methods
6.4. Extraction of knowledge from data
6.5. Neural networks for classifying and grouping data
6.6. Fuzzy-neural and neuron-fuzzy systems
6.7. Special cases of fuzzy-neural network
6.8. Final remarks
Literature

7. Processing granular information in the process of constructing systems

interactive

Witold Pedrycz

7.1. Admission
7.2. The concept of granular computing environment
7.2.1. Collections and interval analysis
7.2.2. Fuzzy sets
7.2.3. Accurate sets
7.2.4. Shaded sets
7.3. Formalization of the computing environment with information grains
7.3.1. Graininess of information
7.3.2. Family of reference information beans
7.3.3. Definition of the granular information environment
7.4. Communication between granular computing environments
7.5. Constructing information grains
7.5.1. An approach based on information obtained from an expert / user
7.5.2. An approach based on experimental data aggregation
7.6. Grainy models
7.7. Final remarks
Literature

Part III. Supporting decisions

8. Approximate collection in decision support

Zdzisław Pawlak, Roman Słowiński

8.1. Admission
8.2. Inference from data
8.3. Accurate sets - basic concepts
8.4. Approach of approximate sets based on the relationship of domination
8.4.1. Inductive inference and domain knowledge
8.4.2. Granules of knowledge in the form of domes of domination
8.4.3. Approach approximate approach based on dominance (DRSA)
8.4.4. Induction of classification patterns in the form of decision rules
8.4.5. An example of applying the DRSA approach
8.5. Applications of approximate domains based on dominance
8.6. End
Literature

9. Statistical decisions in system analysis

Olgierd Hryniewicz

9.1. Basic problems of decision making in systemic research
9.2. A classic model in the theory of decision making
9.3. Making decisions as a problem of verification of statistical hypotheses
9.4. Bayesian methods of verification of statistical hypotheses
9.5. Bayesian methods of verification of statistical hypotheses for non-precision data
9.6. Posybilistic approach to Bayesian verification of statistical hypotheses
9.7. Bayesian decisions in the case of imprecise hypotheses
statistical and imprecisely determined loss function
Literature

10. Blurred dynamic programming

Janusz Kacprzyk

10.1. Admission
10.2. Fuzzy sets and fuzzy dynamic systems
10.2.1. Basic elements of the fuzzy sets theory
10.2.2. Deterministic, stochastic and fuzzy dynamical systems
10.2.2.1. Deterministic controlled system
10.2.2.2. Stochastic controlled system
10.2.2.3. Fuzzy controlled system
10.3. Multistage decision making and control in the conditions of fuzzyness
10.3.1. Making decisions in a fuzzy environment - Bellman's and Zadeh's approach
10.3.2. Multi-stage decision making (control) in a fuzzy environment
10.4. Control processes with a predetermined end time
10.4.1. Control of the deterministic system
10.4.2. Control of the stochastic system
10.4.2.1. The formulation of Bellman and Zadeh
10.4.2.2. Formulation of Kacprzyk and Staniewski
10.4.3. Control of the fuzzy system
10.5. Remarks about control processes with implicitly specified end time
10.6. Control processes with fuzzy finish time
10.6.1. Control of the deterministic system
10.6.2. Control of the stochastic system
10.6.3. Remarks about controlling the fuzzy system
10.7. Control processes with infinite end time
10.7.1. Control of the deterministic system
10.7.2. Control of the stochastic system
10.7.3. Remarks about controlling the fuzzy system
10.8. Examples of fuzzy dynamic programming applications
Literature

11. Bayesian networks in making decisions

Mieczysław Alojzy Kłopotek

11.1. Introduction
11.1.1. Decision-making processes
11.1.2. Uncertainty in making decisions
11.2. The concept of the Bayesian network
11.2.1. Total probability distribution
11.2.2. Representation of immediate causes
11.2.3. Representation of conditional independence
11.2.4. Types of graphical structures of Bayesian networks
11.2.5. Types of variables in Bayesian networks
11.3. Methods of making decisions in the Bayesian network
11.3.1. Markov tree and inference
11.3.2. Conversion of the Bayesian network into the Markov tree
11.3.3. Special cases of conversion of the Bayesian network into the Markov tree
11.4. Acquisition of Bayesian networks
11.4.1. Teaching Bayesian network with known structure with incomplete data
11.4.2. Learning an unknown network structure
11.5. Applications of Bayesian networks in making decisions
11.6. End
Literature

12. The approach of ordinal regression to multi-criteria ordering

decision variants

Roman Słowiński

12.1. Admission
12.2. Disaggregation of preferential information by ordinal regression - UTA method
12.3. Motivation for proposals for creating a new approach
12.4. A new method of multi-criteria ordering of decision-making variants based
on ordinal regression
12.5. Possible extensions of the proposed method
12.6. Final remarks
Literature

Part IV. Usage

13. Systematic approach to selected issues of telemedicine

Ryszard Tadeusiewicz

13.1. Introduction
13.2. General characteristics of telemedicine
13.3. Areas of telemedicine application
13.4. Technological progress and development of telemedicine
13.5. Remote therapy as a new challenge for telemedicine
13.6. Non-technical and non-medical aspects of telemedicine development
13.7. The role of systems engineering in the development of telemedicine
13.8. Summary
Literature

14. Flexible valuation of network services

Krzysztof Malinowski

14.1. Introduction
14.2. Prices and fixed fees
14.3. Prices and variable fees, tariffs using the concept of effective bandwidth
14.4. Prices and variable fees, flexible contracts on the Internet
14.5. Network as a public infrastructure
Literature

15. Analytical methods and artificial intelligence in damage diagnostics

Józef Korbicz

15.1. Introduction
15.2. Tasks and structure of the diagnostic system
15.3. Diagnostic system with analytical models
15.3.1. Residual generation
15.3.1.1. Parity relations
15.3.1.2. Status observers
15.3.1.3. Observer with unknown entrances
15.3.1.4. Parametric identification
15.4. Intelligent calculations in diagnostic systems
15.4.1. Evolutionary algorithms
15.4.2. Artificial neural networks
15.4.3. Neural networks with external dynamics
15.4.4. Neural networks with internal dynamics
15.4.5. Dynamic neural networks of the GMDH type
15.4.6. Logic fuzzy in diagnostics
15.5. Diagnostics of the automation actuator
15.6. End
Literature

16. Applications of immunological algorithms in exploratory analysis

data

Sławomir T. Wierzchoń

16.1. Introduction
16.2. The immune system
16.3. Artificial immune systems
16.3.1. The choice of space shapes
16.3.2. Binding strength of the paratope-epitope
16.3.3. Choosing an immunological algorithm
16.4. Immunological machine learning algorithms
16.4.1. Unattended machine learning
16.4.2. Supervised machine learning
16.4.3. High-dimensional data representation
16.5. Summary

Literature

Index