SLIDING WINDOW BASED FREQUENT PATTERNS MANAGEMENT METHOD FOR MINING WEIGHTED MAXIMAL FREQUENT PATTERNS OVER DATA STREAM

摘要

A sliding window-based frequent pattern management method comprises the following steps: (a) scanning whether a transaction is collected to the size of a particular batch; (b) inserting a sliding window in the transaction collected to the size of the particular batch; (c) updating a tree data structure based on the transaction inserted in the sliding window; and (d) rebuilding the tree data structure to align a support level on each item included in the tree data structure in a descending order.