Logics for Representing Data Mining Tasks in Inductive Databases

Hong-Cheu Liu, Millist W. Vincent, Jixue Liu, Jiuyong Li

Abstract: We present a logical framework for querying inductive databases, which can accommodate a variety of data mining tasks, such as classification, clustering, finding frequent patterns and outliers detection. We also address the important issues of the expressive power of inductive query languages. We show that the proposed logic programming paradigm has equivalent expressive power to an algebra for data mining presented in the literature [1].