Practical Applications of Data Mining

2.6 Mining of CoMpaCt rules 61

2.6.3 Learning Process

The learning process first begins with the recording of the tuple numbers for

the same attribute value of a single attribute and for each combination of at-

tribute values of two different attributes by scanning the generalized relation

once. For example, the single-attribute subtables for the two attributes MPG

(miles per gallon) and weight can be created as shown in Figure 2.21.

The single-attribute subtables can be expanded further with more attri-

butes. For example, consider the three attributes MPG, weight, and horsepower.

Two-attribute subtables involving these three attributes ...