Contributions of Domain Knowledge and Stacked Generalization in AI-Based Classification Models
We exploit the merits of C4.5 decision tree classifier with two stacking meta-learners: back-propagation multilayer perceptron neural network and naive-Bayes respectively. The performance of these two hybrid classification schemes have been empirically tested and compared with C4.5 decision tree using two US data sets (raw data set and new data set incorporated with domain knowledge) simultaneously to predict US bank failure. Significant improvements in prediction accuracy and training efficiency have been achieved in the schemes based on new data set. The empirical test results suggest that the proposed hybrid schemes perform marginally better in term of AUC criterion.
KeywordsDomain Knowledge True Positive Rate Positive Instance Negative Instance Bank Failure
Unable to display preview. Download preview PDF.
- 3.Fawcett, T.: ROC Graphs: Notes and Practical Considerations for Researchers (2004), http://www.hpl.hp.com/personal/Tom_Fawcett/papers/ROC101.pdf
- 4.George, H.H., Donald, G.S., Alan, B.C.: Bank management: text and cases. John Wiley & Sons, Inc, Chichester (1994)Google Scholar
- 5.Hirsh, H., Noordewier, M.: Using background knowledge to improve inductive learning of DAS sequences. In: Proceedings of IEEE Conference on AI for Applications (1994)Google Scholar
- 6.John, G., Kohavi, R., Pfleger, K.: Irrelevant features and subset selection problem. In: Proceedings of 11th International Conference on Machine Learning (1994)Google Scholar
- 7.Koller, D., Sahami, M.: Toward optimal feature selection. In: Proceedings of the 13th International Conference on Machine Learning (1996)Google Scholar
- 8.Ledezma, A., Aler, R., Borrajo, D.: Empirical study of a stacking state-space - Tools with Artificial Intelligence. In: Proceedings of the 13th International Conference. IEEE Expert, vol. 7-9, pp. 210–217 (2001)Google Scholar
- 11.Witten, I.H., Frank, E.: Data mining—Practical machine learning tools and techniques with Java implementation. Morgan Kaufmann Publisher, San Francisco (1999)Google Scholar