数据挖掘（4）：不同的分类模型有关金融数据分类的评价（accuracy、precision，recall和F1-score）_xgbclassifier 多分类 模型评价

1.首先读入数据（已经处理（删除，填补，类型转换，归一化）过的数据）并定义计算accuracy、precision，recall和F1-score的函数，并对数据采用sklearn中的特征递归消除法进行特征选择

from sklearn import metrics
import pandas as pd
from sklearn.feature_selection import RFE
 
def metrics_result(true, predict):#定义并输出正确率、精确率、召回率、AUC
    acc = metrics.accuracy_score(true, predict)
    pre = metrics.precision_score(true, predict)
    reca = metrics.recall_score(true, predict)
    f_sco = metrics.f1_score(true, predict)
#    auc_ = metrics.auc(true, predict)
    return acc, pre, reca, f_sco
 
 
data = pd.read_csv('data_imp.csv')#读入已经处理（删除，填补，类型转换，归一化）过的数据
label = data['status']
data = data.iloc[:,:-1]
 
data_=RFE(estimator=RandomForestClassifier(), n_features_to_select=30).fit_transform(data,label)

2.定义相关的分类函数

from sklearn.model_selection import StratifiedKFold
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn import svm
from xgboost.sklearn import XGBClassifier
 
def LR_classifier(train_data, train_label, test_data, test_label):
    clf = LogisticRegression(C=1.0, max_iter=1000)
    prediction_test = clf.fit(train_data, train_label).predict(test_data)
    prediction_train = clf.fit(train_data, train_label).predict(train_data)
    return prediction_test, prediction_train
 
def svm_classifier(train_data, train_label, test_data, test_label):
    clf = svm.SVC(C=1.0, kernel='linear', gamma=20)
    prediction_test = clf.fit(train_data, train_label).predict(test_data)
    prediction_train = clf.fit(train_data, train_label).predict(train_data)
    return prediction_test, prediction_train
 
def dt_classifier(train_data, train_label, test_data, test_label):
    clf = DecisionTreeClassifier(max_depth=5)
    prediction_test = clf.fit(train_data, train_label).predict(test_data)
    prediction_train = clf.fit(train_data, train_label).predict(train_data)
    return prediction_test, prediction_train
def rf_classifier(train_data, train_label, test_data, test_label):
    clf = RandomForestClassifier(n_estimators=8, random_state=5, max_depth=6, min_samples_split=2)
    prediction_test = clf.fit(train_data, train_label).predict(test_data)
    prediction_train = clf.fit(train_data, train_label).predict(train_data)
    return prediction_test, prediction_train
def xgb_classifier(train_data, train_label, test_data, test_label):
    clf = XGBClassifier(n_estimators=8,learning_rate= 0.25, max_depth=20,subsample=1,gamma=13, seed=1000,num_class=1)
    prediction_test = clf.fit(train_data, train_label).predict(test_data)
    prediction_train = clf.fit(train_data, train_label).predict(train_data)
    return prediction_test, prediction_train

然后再进行四折交叉验证，最后结果如下：

模型	accuracy	precision	recall	f1_score	auc	roc curve
Logistic Regression	train:0.7909, test:0.7868	train:0.7352, test:0.7209	train:0.2638, test:0.2525	train:0.3883, test:0.3731	train:0.6132, test:0.6216
Support Vector Machine	train:0.778, test:0.7753	train:0.8144, test:0.8023	train:0.1524, test:0.1437	train:0.2568, test:0.2433	train:0.5694, test:0.5743
Decision Tree	train:0.8082, test:0.7735	train:0.766, test:0.6196	train:0.345, test:0.2704	train:0.4734, test:0.3755	train:0.6272, test:0.6138
Random Forest	train:0.8218, test:0.7768	train:0.8931, test:0.661	train:0.3321, test:0.2354	train:0.4839, test:0.3457	train:0.6642, test:0.6119
XGBoost	train:0.8138, test:0.7845	train:0.7897, test:0.6655	train:0.3543, test:0.2875	train:0.4889, test:0.4012	train:0.6610, test:0.6455