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"]]
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 10000.00loan_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 1280.00tot_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 1veh_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()
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()
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()
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()
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()