
49Data Mining Techniques
organized in levels. The nodes in level l have item sets of size of l. For example, the sizes of the
nodes (i.e., the size of the item sets corresponding to these nodes) in levels 1 and 2 are 1 and 2,
respectively. The root of the graph, level 0, is an empty node corresponding to an empty item set. A
node X in level l is linked to a node Y in level l + 1 if X ⊂ Y. To further explain the process, suppose
we have the following sample web transactions involving pages 1, 2, 3, 4, and 5 as in Table 4.2. The
Apriori algorithm produces the item sets as in Table 4.3, using a minsup = 0.49. The Frequent Item
Set Graph is sho