Incremental Induction of Probabilistic Rules Based on Incremental Sampling Scheme

This paper proposes a new framework for incremental learning based on incremental sampling scheme and rule layers constrained by inequalities of accuracy and coverage. Incremental sampling scheme shows that the number of patterns of updates of accuracy and coverage is four, which give two important inequalities of accuracy and coverage for induction of probabilistic rules. By using these two inequalities, the proposed method classifies a set of formulae into three layers: the rule layer, sub rule layer and the non-rule layer. Using these layers, updates of probabilistic rules are equivalent to their movement between layers. The proposed method was evaluated on datasets regarding headaches and meningitis, and the results show that the proposed method outperforms the conventional methods.