Frequent Pattern Mining 107
space. Their algorithm, Moment, uses an in-memory prefix-tree-based struc-
ture, called the Closed Enumeration Tree (CET), to maintain a dynamically
selected set of itemsets over a sliding-window. Let v
X
be a node representing
the itemset X in the CET. The dynamically selected set of itemsets (nodes)
are classified into the following four types.
• Infrequent Gateway Nodes (IGN): v
X
is an IGN if:
– X is infrequent,
– v
Y
is the parent of v
X
and Y is frequent,
– if v
Y
has a sibling, v
Y
0
, such that X = Y ∪Y
0
, then Y
0
is frequent;
• Unpromising Gateway Nodes (UGN): v
X
is a UGN if:
– X is frequent,
– ∃Y such that Y is a frequent closed itemset, Y ⊃ X, freq(Y ) =
freq(X) and Y is before X according to the lexicographical order
of the itemsets; ...