Identifying Corporate Performance Factors Based on Feature Selection in Statistical Pattern Recognition : Methods, Application, Interpretation.

Intro -- Contents -- Introduction -- 1 Formulation of objectives and methodologicalapproach -- 1.1 Summary of previous research activities -- 1.2 Methodology of current research -- 2 Competitiveness and its measurement -- 2.1 The term competitiveness -- 2.2 Approaches to measuring competitiveness -- 2.3 Financial performance indicators used -- 2.4 Period of performance measurement -- 2.5 The development of performance measurementmethodology -- 2.5.1 Cluster analysis -- 2.5.2 Hyperbola -- 2.5.3 Summation -- 2.5.4 Quintiles -- 2.6 Assessing the appropriateness of methods to measurefinancial performance -- 2.6.1 Experiment settings -- 2.6.2 Experiment output -- 2.7 Description of the methodology used to measureperformance -- 3 Feature Selection Methods in Statistical PatternRecognition -- 3.1 Introduction -- 3.1.1 Common Research Issues in Machine Learning and Management -- 3.2 Dimensionality Reduction -- DR Categorization According to Nature of the Resulting Features -- DR Categorization According to the Aim -- 3.3 Feature Subset Selection -- 3.3.1 FS Categorization With Respect to Optimality -- 3.3.2 FS Categorization With Respect to Selection Criteria -- 3.3.3 FS Categorization With Respect to Problem Knowledge -- 3.4 Sub-optimal Search Methods -- 3.4.1 Best Individual Features -- 3.4.2 Sequential Search Methods and their Evolution -- Floating search methods -- Oscillating search method -- 3.4.3 Non-sequential and alternative methods -- 3.4.4 Pitfalls of feature subset evaluation - experimental comparisonof criterion functions -- 3.4.5 Summary of recent sub-optimal feature selection methods -- 3.4.6 Dependency-Aware Feature Selection (DAF) -- 3.5 Performance Estimation Problem -- 3.6 Problem of Feature Selection Overfitting and Stability -- 3.6.1 Problem of Feature Selection Stability -- 3.7 Summary.

4 Testing approaches and methods basedon learning methods for identifying factorsof competitiveness -- 4.1 Introduction -- 4.2 Feature selection based evaluation of competitivenessfactors -- 4.2.1 Feature Selection Methodology -- 4.2.2 Evaluating Stability of Feature Selection Methods -- 4.3 Introducing the modified feature selection methodology -- Non-Parametric Model -- Handling Missing Values and Non-Numeric Values -- 4.4 Pattern classification approach -- 4.5 Regression approach and pseudo-kernel regressionmodel -- 4.6 Experiments and results -- 4.6.1 Regression-based analysis results -- 4.6.2 Classification-based analysis results -- 4.7 Comparing Regression-based and Classification-basedanalysis results -- 4.8 Improved Model for Attribute Selectionon High-Dimensional Economic Data -- 4.8.1 Improvements of the regression model -- 4.8.2 Optimized model performance on 37- and 74-dim data -- 4.9 Conclusions -- 5 Identifying factors of competitiveness usingbivariate analyses and linearregression analyses -- 5.1 General characteristics -- Effect of company size on the financial performance -- Effect of industry on the financial performance -- Interaction of company size and industry -- 5.2 Internal competitiveness factors of a company -- Multidimensional model -- 5.3 External competitiveness factors of a company -- Multidimensional model -- 5.4 Stakeholder orientation and characteristicsof an organizational structure -- Multidimensional model -- 5.5 Owners -- Multidimensional model -- 5.6 Employees -- Multidimensional model -- 5.7 Customers -- Multidimensional model -- 5.8 Suppliers -- Multidimensional model -- 5.9 Corporate social responsibility -- Multidimensional model -- 5.10 Comprehensive model -- 6 Interpretation of the results achieved -- 6.1 Overall characteristics of the sample.

6.2 Identification of typical combinations of factor valuesleading to certain types of financial performance -- Group A as a whole -- Group A 1 -- Group A 2 -- Group A 3 -- Group A 4 -- 6.3 Summary -- 7 Conclusion -- 8 Bibliography -- List of Figures -- List of Tables -- List of Graphs.

Description based on publisher supplied metadata and other sources.

Electronic reproduction. Ann Arbor, Michigan : ProQuest Ebook Central, 2024. Available via World Wide Web. Access may be limited to ProQuest Ebook Central affiliated libraries.