【数据挖掘】关联模式评估方法及Apriori算法超市购物应用实战（超详细 附源码）

import pandas as pd
from apyori import apriori
df = pd.read_excel("data.xls")
df.head()
min_supp = 0.1  
min_conf = 0.1  
min_lift = 0.1  
result = list(apriori(transactions=transactions, min_support=min_supp, 
           min_confidence=min_conf, min_lift=min_lift))
result
supports=[]
confidences=[]
lifts=[]
bases=[]
adds=[]
for r in result:
    for x in r.ordered_statistics:
        supports.append(r.support)
        confidences.append(x.confidence)
        lifts.append(x.lift)
        bases.append(list(x.items_base))
        adds.append(list(x.items_add))
resultshow = pd.DataFrame({
    'support':supports,
    'confidence':confidences,
    'lift':lifts,
    'base':bases,
    'add':adds})

mport pandas as pd
item_list = [['牛奶','面包'],
    ['面包','尿布','啤酒','土豆'],
    ['牛奶','尿布','啤酒','可乐'],
    me(item_list)
# from mlxtend.preprocessing import TransactionEncode
import mlxtend
te = mlxtend.preprocessing.TransactionEncoder()
df_tf = te.fit_transform(item_list)
df = pd.DataFrame(df_tf,columns=te.columns_)
display(df)
计算频繁项集
from mlxtend.frequent_patterns import apriori
# use_colnames=True表示使用元素名字，默认的False使用列名代表元素frequent_itemsets = apriori(df, min_support=0.05, use_colnames=True)
frequent_itemsets.sort_values(by='support', ascending=False, inplace=True)
# 选择2频繁项集
print(frequent_itemsets[frequent_itemsets.itemsets.apply(lambd