can be achieved by rolling up a data cube to the smallest level of aggregation needed, reducing the dimensionality, or dividing continuous measures into discrete intervals.

Data reduction

Data cleansing

Inferencing

Clustering

Normalization