Paulooh007

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

Paulooh007

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

Paulooh007

train = pd.read_csv('train.csv')
test = pd.read_csv('test.csv')
print('train data shape: ',train.shape,'\ntest data shape: ',test.shape)

Paulooh007

# The first step is to convert the data type of the column (TRANSACTION TIME) to DateTime. 
# That can be done using .to_datetime() method in Pandas.
train['transaction time']= pd.to_datetime(train['transaction time'])
train['hour'] = train['transaction time'].dt.hour #returns the hour of the datetime(range 1 - 24)
train['dayofweek_name'] = train['transaction time'].dt.weekday_name #day of week of transaction
train['is_weekend'] = np.where(train['dayofweek_name'].isin(['Sunday','Saturday']),1,0) #weekend or not

#Getting the period of the day from the hour feature
bins = [0,4,8,12,16,20,24]
period = ['Late Night', 'Early Morning','Morning','Noon','Eve','Night']

Paulooh007

#Getting the recent amount withdrawn or deposited
train['prev_tran'] = train['current bank amount'] - train['last bank amount']
#credit if the value is positive or debit if otherwise
train['credit_or_debit'] = ['credit' if x > 0 else 'debit' for x in train['prev_tran'] ]

Paulooh007

features = train[['current bank amount', 'last bank amount',
       'time taken (seconds)', 'most recent bank amount', 'account type',
       'age', 'occupation', 'credit card type',
       'account source verification', 'transaction source method',
       'account destination verification', 'dayofweek_name', 'is_weekend',
       'hour','prev_tran','credit_or_debit','period_of_day']]
X = pd.get_dummies(columns=['account type',
       'occupation', 'credit card type', 'credit_or_debit',
       'account source verification', 'transaction source method',
       'account destination verification', 'dayofweek_name','period_of_day'], data = features)

Paulooh007

from sklearn import model_selection
from sklearn.linear_model import LogisticRegression
from sklearn.tree import DecisionTreeClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
from sklearn.metrics import auc, roc_auc_score

Paulooh007

models = []
models.append(('Lr', LogisticRegression()))
models.append(('LDA', LinearDiscriminantAnalysis()))
models.append(('Knn', KNeighborsClassifier()))
models.append(('Decision_Tree', DecisionTreeClassifier()))

Paulooh007

# evaluate each model in turn
results = []
names = []
for name, model in models:
    model.fit(X_train, y_train)
    score = roc_auc_score(y_test, model.predict_proba(X_test)[:,1])
    results.append(score)
    names.append(name)
    msg = "%s: %f " % (name, score)
    print(msg)

Paulooh007

test['transaction time']= pd.to_datetime(test['transaction time'])
test['hour'] = test['transaction time'].dt.hour #returns the hour of the datetime(range 1 - 24)
test['dayofweek_name'] = test['transaction time'].dt.weekday_name #day of week of transaction
test['is_weekend'] = np.where(test['dayofweek_name'].isin(['Sunday','Saturday']),1,0) #weekend or not

test['period_of_day'] = pd.cut(test['hour'], bins=bins, labels=period)

#Getting the recent amount withdrawn or deposited
test['prev_tran'] = test['current bank amount'] - test['last bank amount']
#credit if the value is positive or debit if otherwise