Table of Contents

Preface

 

1. Knowledge Discovery from Data

1.1 Models and their application in decision making

1.2 Relevance and value of forecasts

1.3 Theory driven approach

1.4 Data driven approach

1.5 Data mining

References

 

2. Self-organising Data Mining

2.1 Involvement of users in the data mining process

2.2 Automatic model generation

2.3 Self-organising data mining

References

 

3. Self-organising Modelling Technologies

3.1 Statistical Learning Networks

3.2 Inductive approach - The GMDH algorithm

References

 

4. Parametric GMDH Algorithms

4.1 Elementary models (neurons)

4.2 Generation of alternate model variants

4.3 Nets of active neurons

4.4 Criteria of model selection

4.5 Validation

References

 

5. Nonparametric Algorithms

5.1 Objective Cluster Analysis

5.2 Analog Complexing

5.3 Self-organising Fuzzy Rule Induction

5.4 Logic based rules

References

 

6. Application of Self-organising Data Mining

6.1 Spectrum of self-organising data mining methods

6.2 Choice of appropriate modelling methods

6.3 Application fields

6.4 Synthesis

6.5 Software tools

References

 

7. KnowledgeMiner

7.1 General features

7.2 GMDH implementation

7.3 Analog Complexing implementation

7.4 Fuzzy Rule Induction implementation

7.5 Using models

 

8. Sample Applications

8.1 ... From Economics

• national economy
• stock prediction
• balance sheet

• sales prediction
• solvency checking
• energy consumption

 

8.2 ... From Ecology

• water pollution
• water quality

 

8.3 ... From other Fields

• heart disease
• U.S. congressional voting behavior

 

References