194 ◾ Corlane Barclay
likely to be fraudulent or an insurance company identifying customers likely to
have high number of claims.
Descriptive Methods
Clustering consists of grouping and collecting a set of objects into similar classes.
Clustering categories can be mutually exclusive and exhaustive or consist of a
richer representation, such as hierarchical or overlapping categories (Fayyad
1996a). e aim is to identify a finite set of categories or clusters to describe the
data (Jain and Dubes 1988). Examples include identifying or discovering simi-
lar clusters, subpopulations of telecommunication, and banking or supermarket
customers.
Summarization involves methods for nding a compact description for a
subset of data such as tabulating the mea ...