FP-Growth algorithm or FP-tree, which retains the item set association information. The FP-tree is mined by starting from each frequent length 1 pattern constructing its conditional pattern base then constructing its conditional FP-tree, and performing mining recursively on such a tree. The patterned growth is achieved by the concatenation of the suffix pattern of the frequent patterns generated from a conditional FP-tree.
FP-Growth algorithm can be explained as follows. The first peruse of the database acquires a list of frequent items in which items are ordered by frequency descending order. This learner efficiently calculates all frequent item sets from the given data. It calculates all frequent items sets from a data set by building the FP-Tree data structure of the transaction database. This is a very compressed copy of the data which in many cases fit into main memory even for large databases. From this FP-Tree, all frequent item sets are derived. A major advantage of FP-Growth compared to Apriori is that it uses only much data consider and is therefore often appropriate even on large data sets.