Mining database semantic relationships

With the explosive growth of the size of databases and the amount of data in them, the discovery of semantic relationships in databases has become a solution to facilitate the integration of those seemingly disparate and autonomous components since it is difficult to detect semantic heterogeneity among databases. An earlier paper provided an approach which uses logical reasoning and object-oriented techniques to discover new semantic relationships of the object classes in multiple databases. However, several prior information are required in the logical reasoning-based knowledge discovery approach. In this paper, an extension of the data mining approach for the discovery of the semantic relationships is proposed. The extended data mining approach includes the original logical reasoning approach with an addition of an association rule mining algorithm.