PAMI_PyKit stands for PAttern MIning Python Kit. It contains a set of python libraries to discover user interest-based patterns in very large databases. The python programs in this kit are organized in the following topics:
-
Traditional algorithms - Conventional algorithms whose input exists in the form of transactional databases (or files).
-
Parallel algorithms - Parallel pattern mining algorithms based on Map-Reduce framework.
- frequentPatternGrowth (FPgrowth)
- multipleSupportFrequentPatternGrowth using user specified minimum item supports (MSFPgrowth)
- multipleSupportFrequentPatternGrowth using percentage based function (IMSFPgrowth)
- correlatedPatternGrowth (CPgrowth)
- periodicFrequentPattern-growth (PFPgrowth)
- periodicFrequentPattern-growth with periodic summaries (PSgrowth)
- ITL-growth
- periodicFrequentPattern-growth using greedy search (PFPgrowthGS)
- periodicFrequentPattern-growth using multiple minimum supports and maximum periodicities (MSPFP-growth)
- periodicFrequentPattern-growth using periodic-ratio (PFPgrowthPR)
- partialPeriodicFrequentPattern-growth using period-support (PPFPgrowth)
- partialPeriodicFrequentPattern-growth using multiple period-supports (PPFPgrowth_MPS)
- recurringPattern-growth (RP-growth)
- HighUtilityItemsetMining (EFIM)
- HighUtilityFrequentItemsetMining (EFIMpp)
- SpatialHighUtilityItemsetMining
- PartialPeriodicSpatialPatternMining ()
- parallelFrequentPatternGrowth
- parallel PeriodicFrequentPatternGrowth