Search

Abstract: Ensemble learning is a method to improve the performance of classification and prediction algorithms. Many studies have demonstrated that ensemble learning can decrease the generalization error and improve the performance of individual classifiers and predictors. However, its performance can be degraded due to multicollinearity problem where multiple classifiers of an ensemble are highly correlated with. This paper proposes a genetic algorithm-based coverage optimization technique in the purpose of resolving multicollinearity problem. Empirical results with bankruptcy prediction on Korea firms indicate that the proposed coverage optimization algorithm can help to design a diverse and highly accurate classification system. [Copyright &y& Elsevier]

Abstract: Multiple classifier combination is a technique that combines the decisions of different classifiers. Combination can reduce the variance of estimation errors and improve the overall classification accuracy. However, direct application of combination schemes developed for pattern recognition to solving business problems has some limitations, because business problems cannot be explained completely by the results provided by machine-learning-driven classifiers alone owing to their intrinsic complexity. To solve such problems, this paper proposes an approach that is capable of incorporating the subjective problem-solving knowledge of humans into the results of quantitative models. Genetic algorithms (GAs) are used to combine classifiers stemming from machine learning, experts, and users. We use our GA-based method to predict the Korea stock price index (KOSPI). [Copyright &y& Elsevier]

Numerous studies on bankruptcy prediction have widely applied data mining techniques to finding out the useful knowledge automatically from financial databases, while few studies have proposed qualitative data mining approaches capable of eliciting and representing experts'' problem-solving knowledge from experts'' qualitative decisions. In an actual risk assessment process, the discovery of bankruptcy prediction knowledge from experts is still regarded as an important task because experts'' predictions depend on their subjectivity. This paper proposes a genetic algorithm-based data mining method for discovering bankruptcy decision rules from experts'' qualitative decisions. The results of the experiment show that the genetic algorithm generates the rules which have the higher accuracy and larger coverage than inductive learning methods and neural networks. They also indicate that considerable agreement is achieved between the GA method and experts'' problem-solving knowledge. This means that the proposed method is a suitable tool for eliciting and representing experts'' decision rules and thus it provides effective decision supports for solving bankruptcy prediction problems. [Copyright &y& Elsevier]