Resampling-and-Classification-Using-SMOTE

#import library
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import MinMaxScaler
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import confusion_matrix
from sklearn.metrics import roc_auc_score
from sklearn.metrics import recall_score, precision_score, f1_score
from sklearn.metrics import roc_curve, roc_auc_score
np.random.seed(42)


Step 1. Loading Dataset
dataset = pd.read_csv('ya.csv',delimiter = ';',names=['id','account_age',
'no_follower','no_following',
'no_userfavourites','no_lists','no_tweets','no_retweet
s','no_hashtags','no_usermention','no_urls','no_char','
no_digits','Class'])

##Pemilihan Data
data = dataset.drop(['id','Class'], axis=1).values
label = dataset.drop(['id','account_age','no_follower','no_following',
'no_userfavourites','no_lists','no_tweets','no_retweet
s','no_hashtags','no_usermention','no_urls','no_char','
no_digits'], axis=1).values


Step 2. Resampling
from imblearn.over_sampling import SMOTE
smote = SMOTE(k_neighbors = 5,ratio=0.25)83
data_smote, label_smote = smote.fit_sample(data, label)


Step 3. Split Data
data_smote_train,data_smote_test,
label_smote_train,label_smote_test=train_test_split(data_
smote,label_smote,test_size=0.2,random_state=42)


Step 4. Normalisasi
mm = MinMaxScaler()
data_smote_train=mm.fit_transform(data_smote_train)
data_smote_test=mm.transform(data_smote_test)


Step 5. Klasifikasi
##Random Forest
clf = RandomForestClassifier (bootstrap=True, criterion='gini',
max_features='auto', min_samples_split = 3,
n_estimators=50,oob_score=True, random_state = 42)

##Fitting Data
clf.fit(data_smote_train, label_smote_train)

##Training predictions
train_smote_data_clf_predictions = clf.predict(data_smote_train)
train_smote_data_clf_probs = clf.predict_proba(data_smote_train)[:, 0]

##Testing predictions
clf_predictions = clf.predict(data_smote_test)
clf_probs = clf.predict_proba(data_smote_test)[:, 0]

##Predicting the Test set results
label_smote_pred = clf.predict(data_smote_test)84


Step 6. Performansi
cm = confusion_matrix(label_smote_test, label_smote_pred)
recall = recall_score(label_smote_test,label_smote_pred)
ps = precision_score(label_smote_test,label_smote_pred)
f1 = f1_score(label_smote_test,label_smote_pred)
auc = roc_auc_score(label_smote_test,label_smote_pred)
clf.oob_score = clf.score(data_smote_test, label_smote_test)
oob_error = 1 - clf.oob_score
fpr, tpr, thresholds = roc_curve(label_smote_test,label_smote_pred)