Discovering Association Rules in Semi-structured Data Sets
Manuel Montes-y-Gómez, Alexander F. Gelbukh, Aurelio López‐López · 2003
Abstract. The discovery of association rules is one of the classic problems of data mining. Typically, it is done over well-structured data, such as databases. In this paper, we present a method of discovery of association rules in semi-structured data, namely, in a set of conceptual graphs. The method is based on conceptual clustering of the data and constructing of a conceptual hierarchy. A feature of the method is the possibility of using different levels of generalization. 1.