Incremental Rules Induction Based on Rule Layers
This paper proposes a new framework for incremental learning based on accuracy and coverage. Classification of addition of example into four cases gives two inequalities for accuracy and coverage. The proposed method classifies a set of formulae into three layers: rule layer, subrule layer and non-rule layer by using the inequalities obtained. Then, subrule layer plays a central role in updating rules. The proposed method was evaluated on a dataset on meningitis, whose results show that it outperforms other conventional rule induction methods.
Keywordsincremental rule induction rough sets accuracy coverage subrule layer
Unable to display preview. Download preview PDF.
- 2.Cestnik, B., Kononenko, I., Bratko, I.: Assistant 86: A knowledge-elicitation tool for sophisticated users. In: EWSL, pp. 31–45 (1987)Google Scholar
- 3.Quinlan, J.R.: Induction of decision trees. Machine Learning 1(1), 81–106 (1986)Google Scholar
- 4.Clark, P., Niblett, T.: The cn2 induction algorithm. Machine Learning 3 (1989)Google Scholar
- 6.Michalski, R.S., Mozetic, I., Hong, J., Lavrac, N.: The multi-purpose incremental learning system aq15 and its testing application to three medical domains. In: AAAI, pp. 1041–1047 (1986)Google Scholar
- 9.Quinlan, J.: C4.5 - Programs for Machine Learning. Morgan Kaufmann, Palo Alto (1993)Google Scholar