OpAC: A New OLAP Operator Based on a Data Mining Method

For a few years, on-line analysis processing (OLAP) and data mining have known parallel and independent evolutions. Some recent studies have shown the interest of the association of these two fields. Currently, we attend the increase of a more elaborated analysis's need. We think that the idea of coupling OLAP and data mining will be able to fulfill this need. We propose to adopt this coupling in order to create a new operator, OpAC (Operator for Aggregation by Clustering), for multidimensional on-line analysis. The main idea of OpAC consists in using the agglomerative hierarchical clustering to achieve a semantic aggregation on the attributes of a data cube dimension.