Apriori算法菜篮子分析

菜篮子分析

import numpy as np
import pandas as pd
data=pd.read_csv("C:\\Users\\Administrator\\Desktop\\heyang_hanyu.csv",header=None,encoding="gbk")
data.head()

trans = []#转化为数组
for i in range(0,500):
    trans.append([str(data.values[i,j]) for j in range(0, 16)])
trans = np.array(trans)
print(trans.shape)

import pandas as pd
from mlxtend.preprocessing import TransactionEncoder

te = TransactionEncoder()
data = te.fit_transform(trans)
data = pd.DataFrame(data, columns = te.columns_)
data.shape

df=data.drop(["nan"],axis=1)#删除列名为nan的整列

df.head()
from mlxtend.frequent_patterns import apriori

#Now, let us return the items and itemsets with at least 0.5% support:
df1=apriori(df, min_support = 0.005, use_colnames = True)

df1.sort_values(["support"], ascending=False)#按降序排列

frequent_itemsets = apriori(df, min_support = 0.005, use_colnames=True)
frequent_itemsets['length'] = frequent_itemsets['itemsets'].apply(lambda x: len(x))
frequent_itemsets

frequent_itemsets["length"].value_counts()#查看长度为1/2/3的有多少个

df2=frequent_itemsets[ (frequent_itemsets['length'] == 2) &
                   (frequent_itemsets['support'] >= 0.005) ]
df2.sort_values(["support"], ascending=False)

frequent_itemsets[ (frequent_itemsets['length'] == 3) &
                   (frequent_itemsets['support'] >= 0.005) ]

from mlxtend.frequent_patterns import association_rules

association_rule = association_rules(frequent_itemsets,metric='confidence',min_threshold=0.9)	# metric可以有很多的度量选项，返回的表列名都可以作为参数
association_rule.sort_values(by='leverage',ascending=False,inplace=True)    #关联规则可以按leverage排序
print(association_rule)

完成