NO.62——100天机器学习实践第五天：用逻辑回归模型分析信用卡欺诈案例

最新推荐文章于 2022-03-08 17:47:22 发布

one named slash

最新推荐文章于 2022-03-08 17:47:22 发布

阅读量331

点赞数

分类专栏：机器学习文章标签：逻辑回归信用卡欺诈

本文链接：https://blog.csdn.net/ghl1390490928/article/details/102473540

版权

机器学习专栏收录该内容

23 篇文章 13 订阅

订阅专栏

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np

%matplotlib inline

#分类计数
count_classes = pd.value_counts(data['Class'], sort = True).sort_index()
count_classes.plot(kind = 'bar')
plt.title("Fraud class histogram")
plt.xlabel("Class")
plt.ylabel("Frequency")

解决样本不均衡：两种方法可以采用——过采样和下采样。过采样是对少的样本（此例中是类别为 1 的样本）再多生成些，使类别为 1 的样本和 0 的样本一样多。而下采样是指，随机选取类别为 0 的样本，是类别为 0 的样本和类别为 1 的样本一样少。下采样的代码如下：

#训练集   loc根据index索引   iloc根据行号索引
X = data.loc[:, data.columns != 'Class']
#特征集
y = data.loc[:, data.columns == 'Class']

# Number of data points in the minority class
number_records_fraud = len(data[data.Class == 1])  #诈骗记录的条数
fraud_indices = np.array(data[data.Class == 1].index)  #诈骗记录的下标

# Picking the indices of the normal classes
normal_indices = np.array(data[data.Class == 0].index)  #正常记录的下标

#下采样    在normal_indices中随机选择number_records_fraud个，replace = False表示没有重复值
random_normal_indices = np.random.choice(normal_indices, number_records_fraud, replace = False)

# 合并 the 2 indices
under_sample_indices = np.concatenate([fraud_indices,random_normal_indices])

# Under sample dataset
under_sample_data = data.loc[under_sample_indices,:]

X_undersample = under_sample_data.loc[:, under_sample_data.columns != 'Class']
y_undersample = under_sample_data.loc[:, under_sample_data.columns == 'Class']

# Showing ratio
print("Percentage of normal transactions: ", len(under_sample_data[under_sample_data.Class == 0])/len(under_sample_data))
print("Percentage of fraud transactions: ", len(under_sample_data[under_sample_data.Class == 1])/len(under_sample_data))
print("Total number of transactions in resampled data: ", len(under_sample_data))

Percentage of normal transactions:  0.5
Percentage of fraud transactions:  0.5
Total number of transactions in resampled data:  984

from sklearn.model_selection import train_test_split

# Whole dataset   测试集占比0.3
X_train, X_test, y_train, y_test = train_test_split(X,y,test_size = 0.3, random_state = 0)

print("Number transactions train dataset: ", len(X_train))
print("Number transactions test dataset: ", len(X_test))
print("Total number of transactions: ", len(X_train)+len(X_test))

# Undersampled dataset  。之所以划分两种数据集是因为通过下采样样本数较少
X_train_undersample, X_test_undersample, y_train_undersample, y_test_undersample = train_test_split(X_undersample
                                                                                                   ,y_undersample
                                                                                                   ,test_size = 0.3
                                                                                                   ,random_state = 0)
print("")
print("Number transactions train dataset: ", len(X_train_undersample))
print("Number transactions test dataset: ", len(X_test_undersample))
print("Total number of transactions: ", len(X_train_undersample)+len(X_test_undersample))

Number transactions train dataset:  199364
Number transactions test dataset:  85443
Total number of transactions:  284807

Number transactions train dataset:  688
Number transactions test dataset:  296
Total number of transactions:  984

这里引入召回率和准确率的概念。

召回率是针对我们原来的样本而言的，它表示的是样本中的正例有多少被预测正确了。这里会产生两种情况，一种是预测正确，把原来的正类预测成正类(TP)，另一种就是预测错误，把原来的正类预测为负类(FN)。

准确率是针对预测正确的案例，不管是正类预测成正类，还是把负类预测出负类，比上所有样本。

(TP+TN)/(TP+FN+FP+TN)

TP,即 True Positive ，判断成了正例，判断正确；把正例判断为正例了

TN,即 True negative, 判断成了负例，判断正确，这叫去伪；把负例判断为负例了

FP，即False Positive ,判断成了正例，但是判断错了，也就是把负例判断为正例了

FN ,即False negative ，判断成了负例，但是判断错了，也就是把正例判断为负例了。

这里引入正则惩罚项。

模型选择的典型方法就是正则化。正则化是结构风险最小化策略的实现，是在经验风险吗上加上一个正则化项或惩罚项。正则化项一般是模型负责度的单调递增函数，模型越复杂，正则化值就越大。比如正则化相可以是模型参数向量的范数。

       其中，第一项是经验风险，第二项为正则化项，lambda ≥ 0为调整两者之间关系的系数。正则化项可以取不同的形式。例如，回归问题中，损失函数是平方损失正则化项可以是参数向量的L2范数。正则化项也可以是参数辆的L1范数。
       第一项的经验风险较小的模型可能较为复杂（有多个非零参数），这时第二项的模型复杂度会较大。正则化的作用是选择经验风险与模型复杂度同时较小的模型。正则化符合奥卡姆剃刀原理。奥卡姆剃刀原理应用于模型选择时变为以下想法：在所有可能选择的模型中，能够很好地解释数据并且十分简单的才是最好的模型，也就是应该选择的模型。从贝叶斯估计的角度来看，正则化对应于模型的先验概率，可以假设复杂的模型有较小的先验概率，简单的模型有较大的先验概率。
       举一个简单的例子：

以最简单的线性分类为例，假设样本特征为X=[1,1,1,1]，模型1的权重W1=[1,0,0,0]，模型二权重W2=[0.25,0.25,0.25,0.25]，虽然W1X=W2X=1；但是权重W1只关注一个特征（像素点），其余特征点都无效，模型具体、复杂、明显，能识别“正方形棉布材质的彩色手帕”，在训练集上训练完后容易导致过拟合，因为测试数据只要在这个像素点发生些许变化，分类结果就相差很大，而模型2的权重W2关注所有特征（像素点），模型更加简洁均匀、抽象，能识别“方形”，泛化能力强。通过L2正则化惩罚之后，模型1的损失函数会增加λ/8，模型2的损失函数会增加λ/32，显然，模型2更趋向让损失函数值更小。

#Recall = TP/(TP+FN)  TP为推测正确且将正集推测为正，FN为推测错误且将正集推测为负
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import KFold, cross_val_score
from sklearn.metrics import confusion_matrix,recall_score,classification_report

def printing_Kfold_scores(x_train_data,y_train_data):
    fold = KFold(5,shuffle=False) 

    # Different C parameters。逻辑回归的惩罚参数
    c_param_range = [0.01,0.1,1,10,100]
    results_table = pd.DataFrame(index = range(len(c_param_range)), columns = ['C_parameter','Mean recall score'])
    results_table['C_parameter'] = c_param_range

    # the k-fold will give 2 lists: train_indices = indices[0], test_indices = indices[1]
    j = 0
    for c_param in c_param_range:
        print('-------------------------------------------')
        print('C parameter: ', c_param)
        print('-------------------------------------------')
        print('')

        recall_accs = []
        #enumerate() 函数用于将一个可遍历的数据对象(如列表、元组或字符串)组合为一个索引序列，同时列出数据和数据下标，一般用在 for 循环当中。
        for iteration, indices in enumerate(fold.split(x_train_data)):

            #把c_param_range代入到逻辑回归模型中，并使用了l1正则化，小数据集可以使用liblinear'
            lr = LogisticRegression(C = c_param,penalty = 'l1',solver='liblinear')

            #使用indices[0]的数据进行拟合曲线，使用indices[1]的数据进
            #values.ravel()用于将数组降到一维
            lr.fit(x_train_data.iloc[indices[0],:],y_train_data.iloc[indices[0],:].values.ravel())

            # Predict values using the test indices in the training data
            y_pred_undersample = lr.predict(x_train_data.iloc[indices[1],:].values)

            # Calculate the recall score and append it to a list for recall scores representing the current c_parameter
            recall_acc = recall_score(y_train_data.iloc[indices[1],:].values,y_pred_undersample)
            recall_accs.append(recall_acc)
            print('Iteration ', iteration,': recall score = ', recall_acc)

        # The mean value of those recall scores is the metric we want to save and get hold of.
        results_table.loc[j,'Mean recall score'] = np.mean(recall_accs)
        j += 1
        print('')
        print('Mean recall score ', np.mean(recall_accs))
        print('')

    best_c = results_table.loc[results_table['Mean recall score'].astype(float).idxmax()]['C_parameter']
    
    # Finally, we can check which C parameter is the best amongst the chosen.
    print('*********************************************************************************')
    print('Best model to choose from cross validation is with C parameter = ', best_c)
    print('*********************************************************************************')
    
    return best_c

best_c = printing_Kfold_scores(X_train_undersample,y_train_undersample)

-------------------------------------------
C parameter:  0.01
-------------------------------------------

Iteration  0 : recall score =  0.8767123287671232
Iteration  1 : recall score =  0.9041095890410958
Iteration  2 : recall score =  0.9830508474576272
Iteration  3 : recall score =  0.9594594594594594
Iteration  4 : recall score =  0.9393939393939394

Mean recall score  0.932545232823849

-------------------------------------------
C parameter:  0.1
-------------------------------------------

Iteration  0 : recall score =  0.863013698630137
Iteration  1 : recall score =  0.8767123287671232
Iteration  2 : recall score =  0.9830508474576272
Iteration  3 : recall score =  0.9459459459459459
Iteration  4 : recall score =  0.8939393939393939

Mean recall score  0.9125324429480454

-------------------------------------------
C parameter:  1
-------------------------------------------

Iteration  0 : recall score =  0.863013698630137
Iteration  1 : recall score =  0.8904109589041096
Iteration  2 : recall score =  0.9830508474576272
Iteration  3 : recall score =  0.9594594594594594
Iteration  4 : recall score =  0.9090909090909091

Mean recall score  0.9210051747084484

-------------------------------------------
C parameter:  10
-------------------------------------------

Iteration  0 : recall score =  0.863013698630137
Iteration  1 : recall score =  0.8767123287671232
Iteration  2 : recall score =  0.9830508474576272
Iteration  3 : recall score =  0.9459459459459459
Iteration  4 : recall score =  0.9090909090909091

Mean recall score  0.9155627459783485

-------------------------------------------
C parameter:  100
-------------------------------------------

Iteration  0 : recall score =  0.8904109589041096
Iteration  1 : recall score =  0.863013698630137
Iteration  2 : recall score =  0.9830508474576272
Iteration  3 : recall score =  0.9594594594594594
Iteration  4 : recall score =  0.8939393939393939

Mean recall score  0.9179748716781454

*********************************************************************************
Best model to choose from cross validation is with C parameter =  0.01
*********************************************************************************

def plot_confusion_matrix(cm, classes,title='Confusion matrix',cmap=plt.cm.Blues):
    '''这个方法用来输出和画出混淆矩阵的
    '''
    #cm为数据，interpolation='nearest'使用最近邻插值，cmap颜色图谱（colormap), 默认绘制为RGB(A)颜色空间
    plt.imshow(cm,interpolation='nearest',cmap=cmap)
    plt.title(title)
    plt.colorbar()
    tick_marks = np.arange(len(classes))
    #xticks(刻度下标，刻度标签)
    plt.xticks(tick_marks, classes, rotation=0)
    plt.yticks(tick_marks, classes)
    #text()命令可以在任意的位置添加文字
    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")
    #自动紧凑布局
    plt.tight_layout()
    plt.ylabel('True label')
    plt.xlabel('Predicted label')

①使用下采样数据训练与测试

import itertools
lr = LogisticRegression(C = best_c, penalty = 'l1', solver='liblinear')
lr.fit(X_train_undersample,y_train_undersample.values.ravel())
y_pred_undersample = lr.predict(X_test_undersample.values)

# Compute confusion matrix
cnf_matrix = confusion_matrix(y_test_undersample,y_pred_undersample)
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()

②分割不同阈值，使用下采样数据进行训练

lr = LogisticRegression(C = best_c, penalty = 'l1', solver='liblinear')
lr.fit(X_train_undersample,y_train_undersample.values.ravel())
#获得预测样本的概率值
y_pred_undersample_proba = lr.predict_proba(X_test_undersample.values)
#分割阈值
thresholds = [0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9]

plt.figure(figsize=(10,10))

j = 1
for i in thresholds:
    y_test_predictions_high_recall = y_pred_undersample_proba[:,1] > i
    
    plt.subplot(3,3,j)
    j += 1
    
    # Compute confusion matrix
    cnf_matrix = confusion_matrix(y_test_undersample,y_test_predictions_high_recall)
    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]  #0和1的分类
    plot_confusion_matrix(cnf_matrix
                          , classes=class_names
                          , title='Threshold >= %s'%i)

Recall metric in the testing dataset:  1.0
Recall metric in the testing dataset:  1.0
Recall metric in the testing dataset:  1.0
Recall metric in the testing dataset:  0.9863945578231292
Recall metric in the testing dataset:  0.9387755102040817
Recall metric in the testing dataset:  0.891156462585034
Recall metric in the testing dataset:  0.8367346938775511
Recall metric in the testing dataset:  0.7755102040816326
Recall metric in the testing dataset:  0.5850340136054422

过采样策略

#SMOTE是过采样的一种，建议采用这种方法
import pandas as pd
from imblearn.over_sampling import SMOTE
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import confusion_matrix
from sklearn.model_selection import train_test_split
from collections import Counter

credit_cards=pd.read_csv('creditcard.csv')

columns=credit_cards.columns
# The labels are in the last column ('Class'). Simply remove it to obtain features columns
features_columns=columns.delete(len(columns)-1)

features=credit_cards[features_columns]
labels=credit_cards['Class']

#该种方法不需要事先把=0的=1的单独拿出来
oversampler=SMOTE(random_state=0)
os_features,os_labels=oversampler.fit_sample(features_train,labels_train)

os_features = pd.DataFrame(os_features)
os_labels = pd.DataFrame(os_labels)
best_c = printing_Kfold_scores(os_features,os_labels)

-------------------------------------------
C parameter:  0.01
-------------------------------------------

Iteration  0 : recall score =  0.8903225806451613
Iteration  1 : recall score =  0.8947368421052632
Iteration  2 : recall score =  0.968861347792409
Iteration  3 : recall score =  0.9578593332673855
Iteration  4 : recall score =  0.958397907255361

Mean recall score  0.9340356022131161

-------------------------------------------
C parameter:  0.1
-------------------------------------------

Iteration  0 : recall score =  0.8903225806451613
Iteration  1 : recall score =  0.8947368421052632
Iteration  2 : recall score =  0.9702998782781896
Iteration  3 : recall score =  0.9600795770545498
Iteration  4 : recall score =  0.9605631945131401

Mean recall score  0.9352004145192607

-------------------------------------------
C parameter:  1
-------------------------------------------

Iteration  0 : recall score =  0.8903225806451613
Iteration  1 : recall score =  0.8947368421052632
Iteration  2 : recall score =  0.9700564346575191
Iteration  3 : recall score =  0.9602994031720908
Iteration  4 : recall score =  0.9608819423835746

Mean recall score  0.9352594405927219

-------------------------------------------
C parameter:  10
-------------------------------------------

Iteration  0 : recall score =  0.8903225806451613
Iteration  1 : recall score =  0.8947368421052632
Iteration  2 : recall score =  0.970255615801704
Iteration  3 : recall score =  0.9599037161605171
Iteration  4 : recall score =  0.9607940119365582

Mean recall score  0.9352025533298407

-------------------------------------------
C parameter:  100
-------------------------------------------

Iteration  0 : recall score =  0.8903225806451613
Iteration  1 : recall score =  0.8947368421052632
Iteration  2 : recall score =  0.9705433218988603
Iteration  3 : recall score =  0.960321385783845
Iteration  4 : recall score =  0.9605961684307712

Mean recall score  0.9353040597727802

*********************************************************************************
Best model to choose from cross validation is with C parameter =  100.0
*********************************************************************************

lr = LogisticRegression(C = best_c, penalty = 'l1', solver='liblinear')
lr.fit(os_features,os_labels.values.ravel())
y_pred = lr.predict(features_test.values)

# Compute confusion matrix
cnf_matrix = confusion_matrix(labels_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()

one named slash

关注

0
点赞
踩
1

收藏

觉得还不错? 一键收藏
0
评论
NO.62——100天机器学习实践第五天：用逻辑回归模型分析信用卡欺诈案例

import pandas as pdimport matplotlib.pyplot as pltimport numpy as np%matplotlib inline#分类计数count_classes = pd.value_counts(data['Class'], sort = True).sort_index()count_classes.plot(kind =...
复制链接

扫一扫