data_mine

求star！求star！求star！

introduce

In this repository implemente 6 class of Association rule data mining algorithm

1.Apriori (apriori.py)

apriori algorithm

2.Aprioricompress(aprioricompress.py)

transaction compression processing for apriori algorithm

3.Apriorihash(apriorihash.py)

hash method for apriori algorithm

4.Aprioriplus(aprioriplus.py)

transaction compress + dataset compress+hash + apriori

5.Fpgrowth(fpgrowth.py)

fp-growth algorithm

6.Fpgrowthplus(fpgrowthplus.py)

dataset compress + fp_growth

• running progress

• the result of association rule data mining

how to use

• download the repository

  git clone https://github.com/blackAndrechen/data_mine
  

• into this folder

  cd data_mine
  

• write your own code,take apriori algorithm for example ``` from apriori import *

data=[[l1,l2,l3,l4], [l1,l3,l5], [l1,l3,l4]] minsupport=2 minconfident=0.6

apr=Apriori() rulelist=apr.generateR(data,minsupport,minconfident) ```

tips

• if you want use others algorithm,the use method is same,for example

  from fp_growth import *
  fp=Fp_growth()
  rule_list=fp.generate_R(data,min_support,min_confident)
  

• in my code ,i use

  groceries.csv

  and

  药方.xls

  data file,you can try running it ``` filename="groceries.csv" minsupport=25 minconf=0.7

filename="药方.xls"

min_support=600

min_conf=0.9

import os currentpath=os.getcwd() path=currentpath+"/dataset/"+filename

path='/home/czpchen/文档/github/data_mine/dataset/groceries.csv'

data=loaddata(path) apr=Apriori() rulelist=apr.generateR(data,minsupport,minconf)

- if you want use youself dataset,suggest you rewrite a function to read youself dataset,And make sure your data set looks like this.

- if you want save the result of Association rule data

savepath='/home/czpchen/文档/github/datamine/log/groceries_apriori.txt'

saverule(rulelist,save_path) ```

Performance analysis

simple analyse of my dataset

Reference

数据挖掘 第三版