# Importing libraries
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder
from sklearn.svm import SVC
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import GridSearchCV
from sklearn.metrics import classification_report, accuracy_score, ConfusionMatrixDisplay, confusion_matrix, roc_curve, auc
import matplotlib.pyplot as plt

# Read csv file into data variable
data = pd.read_csv('Telco-Customer-Churn.csv')

# Splitting the data into training and testing, using Churn as the label
train_df, test_df = train_test_split(
    data, 
    test_size=0.3, 
    stratify=data['Churn'])

# Displays the top rows of each new data set
display(train_df.head())
display(test_df.head())

	customerID	gender	SeniorCitizen	Partner	Dependents	tenure	PhoneService	MultipleLines	InternetService	OnlineSecurity	...	DeviceProtection	TechSupport	StreamingTV	StreamingMovies	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	TotalCharges	Churn
313	2799-ARNLO	Female	1	Yes	No	5	Yes	No	Fiber optic	No	...	No	No	No	No	Month-to-month	Yes	Electronic check	69.35	341.6	No
911	7508-SMHXL	Female	1	No	No	15	Yes	Yes	Fiber optic	No	...	No	No	Yes	No	Month-to-month	No	Credit card (automatic)	89.00	1288.3	No
5332	4565-EVZMJ	Female	0	No	No	47	Yes	No	DSL	Yes	...	No	Yes	No	No	One year	Yes	Mailed check	58.90	2813.05	No
4062	1915-IOFGU	Female	0	No	No	1	Yes	No	Fiber optic	No	...	No	No	No	No	Month-to-month	No	Electronic check	70.50	70.5	Yes
2629	5028-GZLDO	Male	0	No	No	2	Yes	No	Fiber optic	No	...	No	No	No	No	Month-to-month	No	Credit card (automatic)	70.70	140.7	Yes

5 rows × 21 columns

	customerID	gender	SeniorCitizen	Partner	Dependents	tenure	PhoneService	MultipleLines	InternetService	OnlineSecurity	...	DeviceProtection	TechSupport	StreamingTV	StreamingMovies	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	TotalCharges	Churn
1281	9531-NSBMR	Female	0	No	No	36	Yes	No	No	No internet service	...	No internet service	No internet service	No internet service	No internet service	One year	Yes	Bank transfer (automatic)	19.25	679.8	No
3159	5150-ITWWB	Male	1	No	No	3	Yes	Yes	Fiber optic	No	...	No	No	Yes	Yes	Month-to-month	Yes	Electronic check	94.85	335.75	No
7024	7398-LXGYX	Male	0	Yes	No	44	Yes	Yes	Fiber optic	Yes	...	Yes	No	No	No	Month-to-month	Yes	Credit card (automatic)	84.80	3626.35	No
3110	6569-KTMDU	Female	0	No	No	1	Yes	No	No	No internet service	...	No internet service	No internet service	No internet service	No internet service	Month-to-month	No	Mailed check	19.10	19.1	No
1067	7030-NJVDP	Male	0	Yes	No	24	Yes	No	Fiber optic	Yes	...	No	Yes	No	Yes	Month-to-month	Yes	Electronic check	89.25	2210.2	No

5 rows × 21 columns

# Handle missing values
train_df.fillna(value=0, inplace=True)
test_df.fillna(value=0, inplace=True)

# Check missing values have been handled
display("Training set columns:")
display(train_df.isnull().sum())
display("Testing set columns:")
display(test_df.isnull().sum())

'Training set columns:'

customerID          0
gender              0
SeniorCitizen       0
Partner             0
Dependents          0
tenure              0
PhoneService        0
MultipleLines       0
InternetService     0
OnlineSecurity      0
OnlineBackup        0
DeviceProtection    0
TechSupport         0
StreamingTV         0
StreamingMovies     0
Contract            0
PaperlessBilling    0
PaymentMethod       0
MonthlyCharges      0
TotalCharges        0
Churn               0
dtype: int64

'Testing set columns:'

customerID          0
gender              0
SeniorCitizen       0
Partner             0
Dependents          0
tenure              0
PhoneService        0
MultipleLines       0
InternetService     0
OnlineSecurity      0
OnlineBackup        0
DeviceProtection    0
TechSupport         0
StreamingTV         0
StreamingMovies     0
Contract            0
PaperlessBilling    0
PaymentMethod       0
MonthlyCharges      0
TotalCharges        0
Churn               0
dtype: int64

#https://www.geeksforgeeks.org/machine-learning/label-encoding-across-multiple-columns-in-scikit-learn/

# Columns to encode (change to numerical values)
columns_to_encode = ['Churn', 'gender', 'Partner', 'Dependents', 'PhoneService', 'PaperlessBilling', 'MultipleLines', 'InternetService', 'OnlineSecurity',
            'OnlineBackup', 'DeviceProtection', 'TechSupport', 'StreamingTV', 'StreamingMovies', 'Contract', 'PaymentMethod']

# for loop to work through each column and display the object to integer conversions
for column in columns_to_encode:
    label_encoder = LabelEncoder()
    train_df[column] = label_encoder.fit_transform(train_df[column])
    test_df[column] = label_encoder.transform(test_df[column])
    print(column, list(zip(label_encoder.classes_, label_encoder.transform(label_encoder.classes_))))


# display updated training and testing data sets
display(train_df.head())
display(test_df.head())

Churn [('No', 0), ('Yes', 1)]
gender [('Female', 0), ('Male', 1)]
Partner [('No', 0), ('Yes', 1)]
Dependents [('No', 0), ('Yes', 1)]
PhoneService [('No', 0), ('Yes', 1)]
PaperlessBilling [('No', 0), ('Yes', 1)]
MultipleLines [('No', 0), ('No phone service', 1), ('Yes', 2)]
InternetService [('DSL', 0), ('Fiber optic', 1), ('No', 2)]
OnlineSecurity [('No', 0), ('No internet service', 1), ('Yes', 2)]
OnlineBackup [('No', 0), ('No internet service', 1), ('Yes', 2)]
DeviceProtection [('No', 0), ('No internet service', 1), ('Yes', 2)]
TechSupport [('No', 0), ('No internet service', 1), ('Yes', 2)]
StreamingTV [('No', 0), ('No internet service', 1), ('Yes', 2)]
StreamingMovies [('No', 0), ('No internet service', 1), ('Yes', 2)]
Contract [('Month-to-month', 0), ('One year', 1), ('Two year', 2)]
PaymentMethod [('Bank transfer (automatic)', 0), ('Credit card (automatic)', 1), ('Electronic check', 2), ('Mailed check', 3)]

	customerID	gender	SeniorCitizen	Partner	Dependents	tenure	PhoneService	MultipleLines	InternetService	OnlineSecurity	...	DeviceProtection	TechSupport	StreamingTV	StreamingMovies	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	TotalCharges	Churn
313	2799-ARNLO	0	1	1	0	5	1	0	1	0	...	0	0	0	0	0	1	2	69.35	341.6	0
911	7508-SMHXL	0	1	0	0	15	1	2	1	0	...	0	0	2	0	0	0	1	89.00	1288.3	0
5332	4565-EVZMJ	0	0	0	0	47	1	0	0	2	...	0	2	0	0	1	1	3	58.90	2813.05	0
4062	1915-IOFGU	0	0	0	0	1	1	0	1	0	...	0	0	0	0	0	0	2	70.50	70.5	1
2629	5028-GZLDO	1	0	0	0	2	1	0	1	0	...	0	0	0	0	0	0	1	70.70	140.7	1

5 rows × 21 columns

	customerID	gender	SeniorCitizen	Partner	Dependents	tenure	PhoneService	MultipleLines	InternetService	OnlineSecurity	...	DeviceProtection	TechSupport	StreamingTV	StreamingMovies	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	TotalCharges	Churn
1281	9531-NSBMR	0	0	0	0	36	1	0	2	1	...	1	1	1	1	1	1	0	19.25	679.8	0
3159	5150-ITWWB	1	1	0	0	3	1	2	1	0	...	0	0	2	2	0	1	2	94.85	335.75	0
7024	7398-LXGYX	1	0	1	0	44	1	2	1	2	...	2	0	0	0	0	1	1	84.80	3626.35	0
3110	6569-KTMDU	0	0	0	0	1	1	0	2	1	...	1	1	1	1	0	0	3	19.10	19.1	0
1067	7030-NJVDP	1	0	1	0	24	1	0	1	2	...	0	2	0	2	0	1	2	89.25	2210.2	0

5 rows × 21 columns

# Set features and target (x and y axis) for training data
x_train = train_df.drop(['Churn', 'customerID', 'TotalCharges', 'MonthlyCharges'], axis=1)  
y_train = train_df['Churn']

# Set features and target (x and y axis) for testing data
x_test = test_df.drop(['Churn', 'customerID',  'TotalCharges', 'MonthlyCharges'], axis=1)
y_test = test_df['Churn']

# Logistic Regression Hyperparameters
lr_model = LogisticRegression(max_iter=1000)


lr_parameters = {
    'C': [0.01, 0.1, 1, 10],
    'penalty': ['l2'], 
    'solver': ['liblinear', 'lbfgs', 'saga'],
    'class_weight': [None, 'balanced']
}

lr_grid = GridSearchCV(lr_model, param_grid=lr_parameters, scoring='accuracy')

lr_grid.fit(x_train, y_train)
print("Best Logistic Regression Parameters:", lr_grid.best_params_)

Best Logistic Regression Parameters: {'C': 0.1, 'class_weight': None, 'penalty': 'l2', 'solver': 'liblinear'}

# SVM Hyperparameters
# https://www.geeksforgeeks.org/machine-learning/svm-hyperparameter-tuning-using-gridsearchcv-ml/
"""
#commented out due to the length of time taken to execute

svm_model = SVC(probability=True)

svm_parameters = [
    {'kernel': ['linear'], 'C': [0.01, 0.1, 1, 10]},
    {'kernel': ['rbf', 'poly'], 'C': [0.01, 0.1, 1, 10], 
     'gamma': [1, 0.1, 0.01, 0.001]}
]

svm_grid = GridSearchCV(svm_model, param_grid=svm_parameters, scoring='accuracy', n_jobs=-1) 

svm_grid.fit(x_train, y_train)
print("Best SVM Params:", svm_grid.best_params_)

"""

svm_model = SVC(probability=True, kernel='linear', C=10)
svm_model.fit(x_train, y_train)

SVC(C=10, kernel='linear', probability=True)

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

# LR predict test data using best parameters identified
best_lr = lr_grid.best_estimator_
y_lr_pred = best_lr.predict(x_test)

print("Logistic Regression Model Accuracy Score:", accuracy_score(y_test, y_lr_pred))
print(classification_report(y_test, y_lr_pred))

# SVM predict test data using best parameters identified
y_svm_pred = svm_model.predict(x_test)

print("SVM Model Accuracy Score:", accuracy_score(y_test, y_svm_pred))
print(classification_report(y_test, y_svm_pred))

Logistic Regression Model Accuracy Score: 0.8012304779933743
              precision    recall  f1-score   support

           0       0.83      0.92      0.87      1552
           1       0.68      0.47      0.56       561

    accuracy                           0.80      2113
   macro avg       0.76      0.69      0.71      2113
weighted avg       0.79      0.80      0.79      2113

SVM Model Accuracy Score: 0.79649787032655
              precision    recall  f1-score   support

           0       0.82      0.92      0.87      1552
           1       0.67      0.46      0.54       561

    accuracy                           0.80      2113
   macro avg       0.75      0.69      0.71      2113
weighted avg       0.78      0.80      0.78      2113

# https://www.geeksforgeeks.org/machine-learning/confusion-matrix-machine-learning/
# confusion matrix for LR model

cm = confusion_matrix(y_test, y_lr_pred, labels=[0, 1])
disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=['0 - No', '1 - Yes'])
disp.plot(cmap=plt.cm.Blues)
plt.title('Confusion Matrix for LR where Customer Churn is Predicted', fontsize=15, pad=20)
plt.xlabel('Predicted values')
plt.ylabel('Actual values')
plt.show()

# confusion matrix SVM model
cm = confusion_matrix(y_test, y_svm_pred, labels=[0, 1])
disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=['0 - No', '1 - Yes'])
disp.plot(cmap=plt.cm.Blues)
plt.title('Confusion Matrix for SVM where Customer Churn is Predicted', fontsize=15, pad=20)
plt.xlabel('Predicted values')
plt.ylabel('Actual values')
plt.show()

# plot auc roc curve
# https://www.geeksforgeeks.org/machine-learning/auc-roc-curve/

y_lr_pred = best_lr.predict_proba(x_test)[:, 1]
y_svm_pred = svm_model.predict_proba(x_test)[:, 1]

test_df = pd.DataFrame(
    {'True': y_test, 'LR': y_lr_pred, 'SVM': y_svm_pred})

plt.figure(figsize=(7, 5))

for model in ['LR', 'SVM']:
    fpr, tpr, _ = roc_curve(test_df['True'], test_df[model])
    roc_auc = auc(fpr, tpr)
    plt.plot(fpr, tpr, label=f'{model} (AUC = {roc_auc:.2f})')

plt.plot([0, 1], [0, 1], 'r--', label='Random Guess')

plt.xlabel('False Positive')
plt.ylabel('True Positive')
plt.title('ROC Curve Graph for LR and SVM Model')
plt.legend()
plt.show()