机器学习实例——信用评级全流程实现

最新推荐文章于 2024-02-22 16:49:42 发布

Code进阶狼人

最新推荐文章于 2024-02-22 16:49:42 发布

阅读量3.4k

点赞数 5

分类专栏： Python机器学习与深度学习文章标签：机器学习

本文链接：https://blog.csdn.net/liuzuoping/article/details/103225058

版权

Python机器学习与深度学习专栏收录该内容

37 篇文章 18 订阅

订阅专栏

文章目录

上篇博文我们对汽车违约数据进行了预测与变量筛选
数据挖掘实例——信用评级
现在我们来做一个全流程的机器学习项目
源数据集与源代码可以在GitHub中下载然后本地调试
GitHub地址

读取并选择数据

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import os
accepts = pd.read_csv('accepts.csv')
rejects = pd.read_csv('rejects.csv')

accepts_x = accepts[["tot_derog","age_oldest_tr","rev_util","fico_score","ltv"]]
accepts_y = accepts['bad_ind']
rejects_x = rejects[["tot_derog","age_oldest_tr","rev_util","fico_score","ltv"]]

在这里插入图片描述

定义缺失值替换函数并填补缺失值

def Myfillna_median(df):
    for i in df.columns:
        median = df[i].median()
        df[i].fillna(value=median, inplace=True)
    return df
accepts_x_filled=Myfillna_median(df=accepts_x)
rejects_x_filled=Myfillna_median(df=rejects_x)

标准化数据

from sklearn.preprocessing import Normalizer
accepts_x_norm = pd.DataFrame(Normalizer().fit_transform(accepts_x_filled))
accepts_x_norm.columns = accepts_x_filled.columns

rejects_x_norm = pd.DataFrame(Normalizer().fit_transform(rejects_x_filled))
rejects_x_norm.columns = rejects_x_filled.columns

利用knn模型进行预测，做拒绝推断

from sklearn.neighbors import NearestNeighbors
from sklearn.neighbors import KNeighborsClassifier
neigh = KNeighborsClassifier(n_neighbors=5, weights='distance')
neigh.fit(accepts_x_norm, accepts_y)
rejects['bad_ind'] = neigh.predict(rejects_x_norm)

将审核通过的申请者和未通过的申请者进行合并

rejects_res = rejects[rejects['bad_ind'] == 0].sample(1340)
rejects_res = pd.concat([rejects_res, rejects[rejects['bad_ind'] == 1]], axis = 0)
data = pd.concat([accepts.iloc[:, 2:-1], rejects_res.iloc[:,1:]], axis = 0)

分类变量转换

bankruptcy_dict = {'N':0, 'Y':1}
data.bankruptcy_ind = data.bankruptcy_ind.map(bankruptcy_dict)

处理异常值

year_min = data.vehicle_year.quantile(0.1)
year_max = data.vehicle_year.quantile(0.99)
data.vehicle_year = data.vehicle_year.map(lambda x: year_min if x <= year_min else x)
data.vehicle_year = data.vehicle_year.map(lambda x: year_max if x >= year_max else x)
data.vehicle_year = data.vehicle_year.map(lambda x: 2018 - x)
data.drop(['vehicle_make'], axis = 1, inplace = True)
data_filled=Myfillna_median(df=data)
X = data_filled[['age_oldest_tr', 'bankruptcy_ind', 'down_pyt', 'fico_score',
       'loan_amt', 'loan_term', 'ltv', 'msrp', 'purch_price', 'rev_util',
       'tot_derog', 'tot_income', 'tot_open_tr', 'tot_rev_debt',
       'tot_rev_line', 'tot_rev_tr', 'tot_tr', 'used_ind', 'veh_mileage',
       'vehicle_year']]
y = data_filled['bad_ind']
print(data_filled)

  age_oldest_tr  bad_ind  bankruptcy_ind  down_pyt  fico_score  loan_amt  \ 0              64.0        1             0.0      0.00    
650.0 17200.00 1 240.0 0 0.0 683.54 649.0 19588.54 2 60.0 1 0.0 0.00 613.0 10500.00 3 35.0 1 0.0 3099.00 603.0 10800.00 4 104.0 0 0.0 0.00 764.0 26328.04 … … … … … … … 4171 103.0 1
0.0 2000.00 552.0 19000.00 4183 142.0 1 0.0 0.00 718.0 15121.58 4186 32.0 1 0.0 2136.00 624.0 15000.00 4194 101.0 1 0.0 2429.67 662.0 18700.00 4208 52.0 1 0.0 2503.00 711.0 10000.00
  loan_term    ltv     msrp  purch_price  ...  tot_derog  tot_income  \ 0            36   99.0  17350.0     17200.00  ...       
7.0 6550.00 1 60 99.0 19788.0 19588.54 … 0.0 4666.67 2 60 92.0 11450.0 13595.00 … 7.0 2000.00 3 60 118.0 12100.0 12999.00 … 3.0 1500.00 4 60 122.0 22024.0 26328.04 … 0.0 4144.00 … … … … … … … … 4171 60 86.0 22000.0 21500.00 …
10.0 1775.00 4183 60 123.0 12325.0 15528.15 … 1.0 2916.67 4186 42 97.0 15525.0 17136.00 … 2.0 2750.00 4194 48 100.0 18700.0 21129.67 … 12.0 5000.00 4208 36 83.0 12000.0 14503.00 … 0.0 1280.00
  tot_open_tr  tot_rev_debt  tot_rev_line  tot_rev_tr  tot_tr  used_ind  \ 0             2.0         506.0         500.0         1.0 
9.0 1 1 11.0 34605.0 57241.0 7.0 21.0 0 2 5.0 3076.5 10009.5 3.0 10.0 1 3 5.0 4019.0 5946.0 4.0 10.0 1 4 2.0 0.0 1800.0 0.0 10.0 0 … … … … … … … 4171 8.0 6085.0 4803.0
6.0 27.0 0 4183 7.0 8760.0 10685.0 5.0 13.0 1 4186 4.0 851.0 500.0 1.0 9.0 1 4194 6.0 6195.0 9951.0 3.0 31.0 0 4208 5.0 0.0 2000.0 0.0 2.0 1
  veh_mileage  vehicle_year   0         24000.0          20.0   1            22.0          18.0   2         19600.0          20.0   3         10000.0          21.0   4            14.0          18.0   ...           ...           ...   4171          1.0          18.0   4183     
27975.0 20.0 4186 75000.0 24.0 4194 0.0 18.0 4208 100141.0 24.0

[7533 rows x 21 columns]

利用随机森林填补变量

from sklearn.ensemble import RandomForestClassifier
clf = RandomForestClassifier(max_depth=5, random_state=0)
clf.fit(X,y)

importances = list(clf.feature_importances_)
importances_order = importances.copy()
importances_order.sort(reverse=True)

cols = list(X.columns)
col_top = []
for i in importances_order[:9]:
    col_top.append((i,cols[importances.index(i)]))
col_top

col = [i[1] for i in col_top]
print(col_top)

[(0.2703856544549641, ‘fico_score’), (0.12407590399997459,
‘age_oldest_tr’), (0.11587163522770565, ‘rev_util’),
(0.09324645022037789, ‘ltv’), (0.08054739301439076, ‘tot_derog’),
(0.07645932670619422, ‘tot_rev_line’), (0.06170165695082909,
‘tot_tr’), (0.03363202702336891, ‘tot_rev_debt’),
(0.026431913888006843, ‘purch_price’)]

变量细筛与数据清洗

from woe import WoE
import warnings
warnings.filterwarnings("ignore")
data_filled.head()
iv_c = {}
for i in col:
    try:
        iv_c[i] = WoE(v_type='c').fit(data_filled[i],data_filled['bad_ind']).optimize().iv 
    except:
        print(i)
    
pd.Series(iv_c).sort_values(ascending=False)

rev_util tot_rev_line tot_rev_debt

fico_score 0.464300 age_oldest_tr 0.200235 tot_derog
0.179381 ltv 0.145189 tot_tr 0.107821 purch_price 0.019769 dtype: float64

WOE转换

WOE_c = data_filled[col].apply(lambda col:WoE(v_type='c',qnt_num=5).fit(col,data_filled['bad_ind']).optimize().fit_transform(col,data_filled['bad_ind']))
WOE_c.head()

在这里插入图片描述

划分数据集

from sklearn.model_selection import  train_test_split
X = WOE_c
y = data_filled['bad_ind']
X_train, X_test, y_train, y_test = train_test_split(X,y,test_size = 0.3, random_state = 0)

def plot_confusion_matrix(cm, classes,
                          title='Confusion matrix',
                          cmap=plt.cm.Blues):
    plt.imshow(cm, interpolation='nearest', cmap=cmap)
    plt.title(title)
    plt.colorbar()
    tick_marks = np.arange(len(classes))
    plt.xticks(tick_marks, classes, rotation=0)
    plt.yticks(tick_marks, classes)

    thresh = cm.max() / 2.
    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):
        plt.text(j, i, cm[i, j],
                 horizontalalignment="center",
                 color="white" if cm[i, j] > thresh else "black")

    plt.tight_layout()
    plt.ylabel('True label')
    plt.xlabel('Predicted label')

构建逻辑回归模型，进行违约概率预测

import itertools
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import confusion_matrix,recall_score,classification_report 
lr = LogisticRegression(C = 1, penalty = 'l1')
lr.fit(X_train,y_train.values.ravel())
y_pred = lr.predict(X_test.values)

# Compute confusion matrix
cnf_matrix = confusion_matrix(y_test,y_pred)
np.set_printoptions(precision=2)

print("Recall metric in the testing dataset: ", cnf_matrix[1,1]/(cnf_matrix[1,0]+cnf_matrix[1,1]))

# Plot non-normalized confusion matrix
class_names = [0,1]
plt.figure()
plot_confusion_matrix(cnf_matrix
                      , classes=class_names
                      , title='Confusion matrix')
plt.show()

在这里插入图片描述

加入代价敏感参数，重新计算

import itertools
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import confusion_matrix,recall_score,classification_report 
lr = LogisticRegression(C = 1, penalty = 'l1', class_weight='balanced')
lr.fit(X_train,y_train.values.ravel())
y_pred = lr.predict(X_test.values)

# Compute confusion matrix
cnf_matrix = confusion_matrix(y_test,y_pred)
np.set_printoptions(precision=2)

print("Recall metric in the testing dataset: ", cnf_matrix[1,1]/(cnf_matrix[1,0]+cnf_matrix[1,1]))

# Plot non-normalized confusion matrix
class_names = [0,1]
plt.figure()
plot_confusion_matrix(cnf_matrix
                      , classes=class_names
                      , title='Confusion matrix')
plt.show()

在这里插入图片描述

检验模型

from sklearn.metrics import roc_curve, auc
fpr,tpr,threshold = roc_curve(y_test,y_pred, drop_intermediate=False) ###计算真正率和假正率  
roc_auc = auc(fpr,tpr) ###计算auc的值  
  
plt.figure()  
lw = 2  
plt.figure(figsize=(10,10))  
plt.plot(fpr, tpr, color='darkorange',  
         lw=lw, label='ROC curve (area = %0.2f)' % roc_auc) ###假正率为横坐标，真正率为纵坐标做曲线  
plt.plot([0, 1], [0, 1], color='navy', lw=lw, linestyle='--')  
plt.xlim([0.0, 1.0])  
plt.ylim([0.0, 1.05])  
plt.xlabel('False Positive Rate')  
plt.ylabel('True Positive Rate')  
plt.title('Receiver operating characteristic example')  
plt.legend(loc="lower right")  
plt.show()

在这里插入图片描述

利用sklearn.metrics中的roc_curve算出tpr，fpr作图

fig, ax = plt.subplots()
ax.plot(1 - threshold, tpr, label='tpr') # ks曲线要按照预测概率降序排列，所以需要1-threshold镜像
ax.plot(1 - threshold, fpr, label='fpr')
ax.plot(1 - threshold, tpr-fpr,label='KS')
plt.xlabel('score')
plt.title('KS Curve')
#plt.xticks(np.arange(0,1,0.2), np.arange(1,0,-0.2))
#plt.xticks(np.arange(0,1,0.2), np.arange(score.max(),score.min(),-0.2*(data['反欺诈评分卡总分'].max() - data['反欺诈评分卡总分'].min())))
plt.figure(figsize=(20,20))
legend = ax.legend(loc='upper left', shadow=True, fontsize='x-large')

plt.show()