Similarity of Attributes by External Probes

In data mining, similarity or distance between attributes is one of the central notions. Such a notion can be used to build attribute hierarchies etc. Similarity metrics can be user-de ned, but an important problem is de ning similarity on the basis of data. Several methods based on statistical techniques exist. For dening the similarity between two attributes A and B they typically consider only the values of A and B, not the other attributes. We describe how a similarity notion between attributes can be de ned by considering the values of other attributes. The basic idea is that in a 0/1 relation r, two attributes A and B are similar if the subrelations A=1(r) and B=1(r) are similar. Similarity between the two relations is de ned by considering the marginal frequencies of a selected subset of other attributes. We show that the framework produces natural notions of similarity. Empirical results on the Reuters-21578 document dataset show, for example, how natural classi cations for co...

• Type: Conference
• Year: 1998
• Where: KDD
• Authors: Gautam Das, Heikki Mannila, Pirjo Ronkainen
• Keywords: Attribute Hierarchies Etc, Data Mining, KDD 1998, Similarity Metrics, Similarity Notion