数据分享|Python用Apriori算法关联规则分析亚马逊购买书籍关联推荐客户和网络图可视化

import numpy as np

import pandas as pd



data.head()

data.shape

#执行堆叠的步骤，转换为字符串，包括删除索引

dt2 = pd.DataFrame

dt2 = dt2.reset_index(drop = True)

dt2.nunique()  # 总共有4,999本独特的书籍

#数据集中购买最多的前10本书

top0 = pd.DataFrame(dt2.value_counts(sort= True, ascending=False).head(10))

to10

# 创建条形图 

plt.bar(t0.index, top_10\['Frequency'\])

tdf= t.fit(d2).transform(da2)

ted= t.fit(r).transform(tr)

t_f

tdf= df.astype("int")



t_f

oks = d.DataFrame(tf, columns=e.columns_)

bos.head()

runets = apriori(o2, min\_support=0.01, use\_colnames=True)

feqts

fetes.sort_values( by = \['support'\] ,ascending = False)

rls  =  assoc(fret, metric = "lift", min_threshold = 1)

re.solues('confidence', ascending = False)

ruls.head()

rul = rls\[res\['confidence'\] >= 0.55\]

rue

fig, ax = plt.subplots(figsize = (10,6))

G = x.from\_pandas\_edgelist(ul,source = 'antecedents')

n.draw(A)