HLCM: closed frequent pattern mining in Haskell
This page describes the HLCM package, which allows to mine closed frequent patterns in Haskell.
Given a transaction database whose transactions are set of items, a frequent itemset is a set of items that occur together in more than a given number of transactions.
HLCM is an Haskell implementation of the LCM algorithm by Hiroki Arimura and Takeaki Uno. See here for the original C code of LCM. LCM is the winner of the FIMI'04 competition about frequent itemset miners.
You can find more details about HLCM, as well as a detailed experimental study about its performance and the influence of RTS parameter, in the draft of the paper that we submitted to the 2010 edition of the Haskell Symposium: PDF.
The HLCM package provides both a stand alone program (hlcm) and a library that you can use from your own programs. It also provides a benchmarking program to test different performance settings.
The Haddock documentations, with details on how to use the program, can be found below:
Install: cabal install hlcm
For direct download and other infos, please see the hackageDB page of hlcm.
Maintainer: Alexandre Termier. Feel free to contact me by email if you have any inquiry !
Last update: 2010/06/17. Page in construction.