A Genetic Algorithm-Based Solution for the Problem of Small Disjuncts

摘要: In essence, small disjuncts are rules covering a number of examples. Hence, these usually error-prone, which contributes to decrease in predictive accuracy. The problem is particularly serious because, although each covers few examples, the set can cover large This paper proposes solution discovering accurate small-disjunct based on genetic algorithms. basic idea our method use hybrid decision tree / algorithm approach for classification. More precisely, examples belonging classified by produced decision-tree algorithm, while new designed rules.