数据挖掘项目（一）

1. 导入数据

import pandas as pd
import numpy as np
df = pd.read_csv('data.csv',encoding="gbk")
df.head()

2. 删除无关的特征

data1 = df.drop(['trade_no', 'bank_card_no', 'source', 'Unnamed: 0', 'id_name'], axis=1)
data1.info()

3. 缺省值处理(以下参考88080917代码)

print(df.isnull().sum())
count=0
for i in range(85):
    if df.isnull().sum()[i]>250:
        count=count+1
print(count)
print(max(fd.isnull().sum()))

4. 剔除，填充，合并

data1=data1.drop(['student_feature'], axis=1)
data1.dropna(thresh=70, inplace = True)

data_col=['loans_latest_time', 'latest_query_time', 'reg_preference_for_trad']
data2 = data1[data_col]
data3 = data1.drop(data_col, axis=1)

d