We describe and compare attribute-selection strategies for attribute-oriented generalization (AOG). AOG summarizes the information in a relational database by repeatedly replacing specific attribute values with more general concepts. Several strategies for selecting the next attribute to generalize have been suggested in the literature, but their relative merits have not previously been assessed. Here, we evaluate the usefulness and efficiency of previously proposed and new strategies. Ten different attribute selection strategies for generalization were implemented and tested, with the performance of the strategies evaluated and compared using criteria that consider their ability to efficiently produce interesting results. We use measures of interestingness that consider the structure of the domain-expert defined concept hierarchies that are used to guide generalization. Bsed on the comparison of the experimental results, a strategy that considers the complexity of the concept hierarchies was found to provide efficient and effective guidance towards interesting results.
展开▼
