This project contains a python implementation of the FP-Growth algorithm used for mining frequent itemsets. See http://en.wikipedia.org/wiki/Association_rule_learning#cite_note-fp-growth-17 for more information.
Usage:
$ python fpgrowth.py -o output_file -s support -t treshold
use -h or --help for help and -v or --verbose for verbose
Default config:
- output_file = fpgrowth-results.out
- support = 3
- treshold = 0.20
Input data:
Input data must follow the schema:
Transaction ID<TAB>Timestamp<TAB>Item<TAB>Item count