In data mining, we emphasize the need for learning from huge, incomplete, and imperfect data sets. To handle noise in the problem domain, existing learning systems avoid overfitting the imperfect training examples by ex- clouding insignificant patterns. The problem is that these systems use a limiting attribute-value language for rep- resenting the training examples and the induced knowl- edge. Moreover, some important patterns are ignored because they are statistically insignificant. In this article, we present a framework that combines Genetic Pro- gramming and Inductive Logic Programming to induce knowledge represented in various knowledge represen- tation formalisms from noisy databases.
展开▼