What is customer churn?¶

Customer churn is defined as when customers or subscribers discontinue doing business with a firm or service.

Customers in the telecom industry can choose from a variety of service providers and actively switch from one to the next. The telecommunications business has an annual churn rate of 15-25 percent in this highly competitive market.

Individualized customer retention is tough because most firms have a large number of customers and can't afford to devote much time to each of them. The costs would be too great, outweighing the additional revenue. However, if a corporation could forecast which customers are likely to leave ahead of time, it could focus customer retention efforts only on these "high risk" clients. The ultimate goal is to expand its coverage area and retrieve more customers loyalty. The core to succeed in this market lies in the customer itself.

Customer churn is a critical metric because it is much less expensive to retain existing customers than it is to acquire new customers.

To detect early signs of potential churn, one must first develop a holistic view of the customers and their interactions across numerous channels.As a result, by addressing churn, these businesses may not only preserve their market position, but also grow and thrive. More customers they have in their network, the lower the cost of initiation and the larger the profit. As a result, the company's key focus for success is reducing client attrition and implementing effective retention strategy.

Objectives:¶

Finding the % of Churn Customers and customers that keep in with the active services.
Analysing the data in terms of various features responsible for customer Churn
Finding a most suited machine learning model for correct classification of Churn and Non-Churn customers.

Dataset:¶

Telco Customer Churn

Exploratory Data Analysis¶

In [1]:

# Standard libraries for data analysis
import pandas as pd
import numpy as np

#Standard libraries for data visualization:
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.express as px
import plotly.graph_objects as go
from plotly.subplots import make_subplots
import warnings
warnings.filterwarnings('ignore')

# Standard libraries for encoding categorical variables
from sklearn.preprocessing import LabelEncoder
from sklearn.preprocessing import StandardScaler

# Libraries for machine learning
from sklearn.neighbors import KNeighborsClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.naive_bayes import GaussianNB
from sklearn.svm import SVC
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
from sklearn import metrics
from sklearn.metrics import roc_curve
from sklearn.metrics import recall_score, confusion_matrix, precision_score, f1_score, accuracy_score, classification_report

In [2]:

# Standard librabries for measuring performance
from sklearn.metrics import confusion_matrix, accuracy_score 
from sklearn.metrics import f1_score, precision_score, recall_score, fbeta_score
from sklearn.model_selection import cross_val_score
from sklearn.model_selection import KFold
from sklearn import model_selection
from sklearn import metrics
from sklearn.metrics import classification_report, precision_recall_curve
from sklearn.metrics import auc, roc_auc_score, roc_curve
from sklearn.metrics import recall_score

In [3]:

# Loading the dataset
data = pd.read_csv("data.csv")
data.head()

Out[3]:

	customerID	gender	Partner	Dependents	tenure	PhoneService	MultipleLines	InternetService	OnlineSecurity	...	DeviceProtection	TechSupport	StreamingTV	StreamingMovies	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	TotalCharges	Churn
0	7590-VHVEG	Female	Yes	No	1	No	No phone service	DSL	No	...	No	No	No	No	Month-to-month	Yes	Electronic check	29.85	29.85	No
1	5575-GNVDE	Male	No	No	34	Yes	No	DSL	Yes	...	Yes	No	No	No	One year	No	Mailed check	56.95	1889.5	No
2	3668-QPYBK	Male	No	No	2	Yes	No	DSL	Yes	...	No	No	No	No	Month-to-month	Yes	Mailed check	53.85	108.15	Yes
3	7795-CFOCW	Male	No	No	45	No	No phone service	DSL	Yes	...	Yes	Yes	No	No	One year	No	Bank transfer (automatic)	42.30	1840.75	No
4	9237-HQITU	Female	No	No	2	Yes	No	Fiber optic	No	...	No	No	No	No	Month-to-month	Yes	Electronic check	70.70	151.65	Yes

5 rows × 21 columns

In [4]:

data.isnull().any().any()

Out[4]:

False

In [5]:

data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 7043 entries, 0 to 7042
Data columns (total 21 columns):
 #   Column            Non-Null Count  Dtype  
---  ------            --------------  -----  
 0   customerID        7043 non-null   object 
 1   gender            7043 non-null   object 
 2   SeniorCitizen     7043 non-null   int64  
 3   Partner           7043 non-null   object 
 4   Dependents        7043 non-null   object 
 5   tenure            7043 non-null   int64  
 6   PhoneService      7043 non-null   object 
 7   MultipleLines     7043 non-null   object 
 8   InternetService   7043 non-null   object 
 9   OnlineSecurity    7043 non-null   object 
 10  OnlineBackup      7043 non-null   object 
 11  DeviceProtection  7043 non-null   object 
 12  TechSupport       7043 non-null   object 
 13  StreamingTV       7043 non-null   object 
 14  StreamingMovies   7043 non-null   object 
 15  Contract          7043 non-null   object 
 16  PaperlessBilling  7043 non-null   object 
 17  PaymentMethod     7043 non-null   object 
 18  MonthlyCharges    7043 non-null   float64
 19  TotalCharges      7043 non-null   object 
 20  Churn             7043 non-null   object 
dtypes: float64(1), int64(2), object(18)
memory usage: 1.1+ MB

In [6]:

data.shape

Out[6]:

(7043, 21)

Checking for missing or garbage values¶

In [7]:

# Removing CustomerID as it is meaningless for our analysis as it is unique to each person
data = data.drop(["customerID"], axis = 1)
data.head()

Out[7]:

	gender	Partner	Dependents	tenure	PhoneService	MultipleLines	InternetService	OnlineSecurity	OnlineBackup	DeviceProtection	TechSupport	StreamingTV	StreamingMovies	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	TotalCharges	Churn
0	Female	Yes	No	1	No	No phone service	DSL	No	Yes	No	No	No	No	Month-to-month	Yes	Electronic check	29.85	29.85	No
1	Male	No	No	34	Yes	No	DSL	Yes	No	Yes	No	No	No	One year	No	Mailed check	56.95	1889.5	No
2	Male	No	No	2	Yes	No	DSL	Yes	Yes	No	No	No	No	Month-to-month	Yes	Mailed check	53.85	108.15	Yes
3	Male	No	No	45	No	No phone service	DSL	Yes	No	Yes	Yes	No	No	One year	No	Bank transfer (automatic)	42.30	1840.75	No
4	Female	No	No	2	Yes	No	Fiber optic	No	No	No	No	No	No	Month-to-month	Yes	Electronic check	70.70	151.65	Yes

In [8]:

data[data["TotalCharges"] == ' ']

Out[8]:

	gender	Partner	Dependents	PhoneService	MultipleLines	InternetService	OnlineSecurity	OnlineBackup	DeviceProtection	TechSupport	StreamingTV	StreamingMovies	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	Churn
488	Female	Yes	Yes	No	No phone service	DSL	Yes	No	Yes	Yes	Yes	No	Two year	Yes	Bank transfer (automatic)	52.55	No
753	Male	No	Yes	Yes	No	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	Two year	No	Mailed check	20.25	No
936	Female	Yes	Yes	Yes	No	DSL	Yes	Yes	Yes	No	Yes	Yes	Two year	No	Mailed check	80.85	No
1082	Male	Yes	Yes	Yes	Yes	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	Two year	No	Mailed check	25.75	No
1340	Female	Yes	Yes	No	No phone service	DSL	Yes	Yes	Yes	Yes	Yes	No	Two year	No	Credit card (automatic)	56.05	No
3331	Male	Yes	Yes	Yes	No	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	Two year	No	Mailed check	19.85	No
3826	Male	Yes	Yes	Yes	Yes	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	Two year	No	Mailed check	25.35	No
4380	Female	Yes	Yes	Yes	No	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	Two year	No	Mailed check	20.00	No
5218	Male	Yes	Yes	Yes	No	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	One year	Yes	Mailed check	19.70	No
6670	Female	Yes	Yes	Yes	Yes	DSL	No	Yes	Yes	Yes	Yes	No	Two year	No	Mailed check	73.35	No
6754	Male	No	Yes	Yes	Yes	DSL	Yes	Yes	No	Yes	No	No	Two year	Yes	Bank transfer (automatic)	61.90	No

In [9]:

data['TotalCharges'] = pd.to_numeric(data.TotalCharges, errors='coerce')
data.isnull().sum()

Out[9]:

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        11
Churn                0
dtype: int64

There are 11 records missing in total charges

In [10]:

data[data["tenure"] == 0]

Out[10]:

	gender	Partner	Dependents	PhoneService	MultipleLines	InternetService	OnlineSecurity	OnlineBackup	DeviceProtection	TechSupport	StreamingTV	StreamingMovies	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	TotalCharges	Churn
488	Female	Yes	Yes	No	No phone service	DSL	Yes	No	Yes	Yes	Yes	No	Two year	Yes	Bank transfer (automatic)	52.55	NaN	No
753	Male	No	Yes	Yes	No	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	Two year	No	Mailed check	20.25	NaN	No
936	Female	Yes	Yes	Yes	No	DSL	Yes	Yes	Yes	No	Yes	Yes	Two year	No	Mailed check	80.85	NaN	No
1082	Male	Yes	Yes	Yes	Yes	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	Two year	No	Mailed check	25.75	NaN	No
1340	Female	Yes	Yes	No	No phone service	DSL	Yes	Yes	Yes	Yes	Yes	No	Two year	No	Credit card (automatic)	56.05	NaN	No
3331	Male	Yes	Yes	Yes	No	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	Two year	No	Mailed check	19.85	NaN	No
3826	Male	Yes	Yes	Yes	Yes	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	Two year	No	Mailed check	25.35	NaN	No
4380	Female	Yes	Yes	Yes	No	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	Two year	No	Mailed check	20.00	NaN	No
5218	Male	Yes	Yes	Yes	No	No	No internet service	No internet service	No internet service	No internet service	No internet service	No internet service	One year	Yes	Mailed check	19.70	NaN	No
6670	Female	Yes	Yes	Yes	Yes	DSL	No	Yes	Yes	Yes	Yes	No	Two year	No	Mailed check	73.35	NaN	No
6754	Male	No	Yes	Yes	Yes	DSL	Yes	Yes	No	Yes	No	No	Two year	Yes	Bank transfer (automatic)	61.90	NaN	No

In [11]:

# Dropping newly acquired customers as this may not aid our prediction model
data.drop(labels=data[data["tenure"] == 0].index, axis = 0, inplace = True)

In [12]:

# Imputing missing "Total Charges" with the mean
data.fillna(data["TotalCharges"].mean())

Out[12]:

	gender	SeniorCitizen	Partner	Dependents	tenure	PhoneService	MultipleLines	InternetService	OnlineSecurity	OnlineBackup	DeviceProtection	TechSupport	StreamingTV	StreamingMovies	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	TotalCharges	Churn
0	Female	0	Yes	No	1	No	No phone service	DSL	No	Yes	No	No	No	No	Month-to-month	Yes	Electronic check	29.85	29.85	No
1	Male	0	No	No	34	Yes	No	DSL	Yes	No	Yes	No	No	No	One year	No	Mailed check	56.95	1889.50	No
2	Male	0	No	No	2	Yes	No	DSL	Yes	Yes	No	No	No	No	Month-to-month	Yes	Mailed check	53.85	108.15	Yes
3	Male	0	No	No	45	No	No phone service	DSL	Yes	No	Yes	Yes	No	No	One year	No	Bank transfer (automatic)	42.30	1840.75	No
4	Female	0	No	No	2	Yes	No	Fiber optic	No	No	No	No	No	No	Month-to-month	Yes	Electronic check	70.70	151.65	Yes
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
7038	Male	0	Yes	Yes	24	Yes	Yes	DSL	Yes	No	Yes	Yes	Yes	Yes	One year	Yes	Mailed check	84.80	1990.50	No
7039	Female	0	Yes	Yes	72	Yes	Yes	Fiber optic	No	Yes	Yes	No	Yes	Yes	One year	Yes	Credit card (automatic)	103.20	7362.90	No
7040	Female	0	Yes	Yes	11	No	No phone service	DSL	Yes	No	No	No	No	No	Month-to-month	Yes	Electronic check	29.60	346.45	No
7041	Male	1	Yes	No	4	Yes	Yes	Fiber optic	No	No	No	No	No	No	Month-to-month	Yes	Mailed check	74.40	306.60	Yes
7042	Male	0	No	No	66	Yes	No	Fiber optic	Yes	No	Yes	Yes	Yes	Yes	Two year	Yes	Bank transfer (automatic)	105.65	6844.50	No

7032 rows × 20 columns

In [13]:

# Checking for missing values
data['TotalCharges'] = pd.to_numeric(data.TotalCharges, errors='coerce')
data.isnull().sum()

Out[13]:

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

In [14]:

# Checking for number of unique values in SeniorCitizen attribute
data.SeniorCitizen.unique()

Out[14]:

array([0, 1])

In [15]:

data.SeniorCitizen = data.SeniorCitizen.map({0: "No", 1: "Yes"})
data.head()

Out[15]:

	gender	SeniorCitizen	Partner	Dependents	tenure	PhoneService	MultipleLines	InternetService	OnlineSecurity	OnlineBackup	DeviceProtection	TechSupport	StreamingTV	StreamingMovies	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	TotalCharges	Churn
0	Female	No	Yes	No	1	No	No phone service	DSL	No	Yes	No	No	No	No	Month-to-month	Yes	Electronic check	29.85	29.85	No
1	Male	No	No	No	34	Yes	No	DSL	Yes	No	Yes	No	No	No	One year	No	Mailed check	56.95	1889.50	No
2	Male	No	No	No	2	Yes	No	DSL	Yes	Yes	No	No	No	No	Month-to-month	Yes	Mailed check	53.85	108.15	Yes
3	Male	No	No	No	45	No	No phone service	DSL	Yes	No	Yes	Yes	No	No	One year	No	Bank transfer (automatic)	42.30	1840.75	No
4	Female	No	No	No	2	Yes	No	Fiber optic	No	No	No	No	No	No	Month-to-month	Yes	Electronic check	70.70	151.65	Yes

In [16]:

data.InternetService.describe(include=["object", "bool"])

Out[16]:

count            7032
unique              3
top       Fiber optic
freq             3096
Name: InternetService, dtype: object

EDA¶

In [17]:

# Visualising the Yes/No Churn distribution
type_ = ["No", "yes"]
fig = make_subplots(rows=1, cols=1)

fig.add_trace(go.Pie(labels=type_, values=data['Churn'].value_counts(), name="Churn"))


fig.update_traces(hole=.4, hoverinfo="label+percent+name", textfont_size=16)

fig.update_layout(
    title_text="Churn Distributions",
    annotations=[dict(text='Churn', x=0.5, y=0.5, font_size=20, showarrow=False)])
fig.show()

In [18]:

data.Churn[data.Churn == "No"].groupby(by = data.gender).count()

Out[18]:

gender
Female    2544
Male      2619
Name: Churn, dtype: int64

The distibution is almost equal

In [19]:

data.Churn[data.Churn == "Yes"].groupby(by = data.gender).count()

Out[19]:

gender
Female    939
Male      930
Name: Churn, dtype: int64

The distribution is almost equal

In [20]:

plt.figure(figsize=(6, 6))
labels =["Churn: Yes","Churn:No"]
values = [1869,5163]
labels_gender = ["F","M","F","M"]
sizes_gender = [939,930 , 2544,2619]
colors = ['#ff6666', '#66b3ff']
colors_gender = ['#c2c2f0','#ffb3e6', '#c2c2f0','#ffb3e6']
explode = (0.3,0.3) 
explode_gender = (0.1,0.1,0.1,0.1)
textprops = {"fontsize":15}

plt.pie(values, labels=labels,autopct='%1.1f%%',pctdistance=1.08, labeldistance=0.8,colors=colors, startangle=90,frame=True, explode=explode,radius=10, textprops =textprops, counterclock = True, )
plt.pie(sizes_gender,labels=labels_gender,colors=colors_gender,startangle=90, explode=explode_gender,radius=7, textprops =textprops, counterclock = True, )

centre_circle = plt.Circle((0,0),5,color='black', fc='white',linewidth=0)
fig = plt.gcf()
fig.gca().add_artist(centre_circle)

plt.title('Churn Distribution w.r.t Gender: Male(M), Female(F)', fontsize=15, y=1.1)
 
plt.axis('equal')
plt.tight_layout()
plt.show()

In [21]:

fig = px.histogram(data, x="Churn", color = "Contract", barmode = "group", title = "Customer contract distribution")
fig.update_layout(width=700, height=500, bargap=0.2)
fig.show()

Customers with monthly contract are more likely to churn

In [22]:

labels = data['PaymentMethod'].unique()
values = data['PaymentMethod'].value_counts()

fig = go.Figure(data=[go.Pie(labels=labels, values=values, hole=.3)])
fig.update_layout(title_text="Payment Method Distribution")
fig.show()

fig = px.histogram(data, x="Churn", color="PaymentMethod", title="Customer Payment Method distribution w.r.t. Churn")
fig.update_layout(width=700, height=500, bargap=0.1)
fig.show()

In [23]:

fig = go.Figure()

fig.add_trace(go.Bar(
  x = [['Churn:No', 'Churn:No', 'Churn:Yes', 'Churn:Yes'],
       ["Female", "Male", "Female", "Male"]],
  y = [965, 992, 219, 240],
  name = 'DSL',
))

fig.add_trace(go.Bar(
  x = [['Churn:No', 'Churn:No', 'Churn:Yes', 'Churn:Yes'],
       ["Female", "Male", "Female", "Male"]],
  y = [889, 910, 664, 633],
  name = 'Fiber optic',
))

fig.add_trace(go.Bar(
  x = [['Churn:No', 'Churn:No', 'Churn:Yes', 'Churn:Yes'],
       ["Female", "Male", "Female", "Male"]],
  y = [690, 717, 56, 57],
  name = 'No Internet',
))

fig.update_layout(title_text="Churn Distribution w.r.t. Internet Service and Gender")

fig.show()

Customers with a fiber optic subscription are more likely to churn when compared to other internet services

In [24]:

data.head()

Out[24]:

	gender	SeniorCitizen	Partner	Dependents	tenure	PhoneService	MultipleLines	InternetService	OnlineSecurity	OnlineBackup	DeviceProtection	TechSupport	StreamingTV	StreamingMovies	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	TotalCharges	Churn
0	Female	No	Yes	No	1	No	No phone service	DSL	No	Yes	No	No	No	No	Month-to-month	Yes	Electronic check	29.85	29.85	No
1	Male	No	No	No	34	Yes	No	DSL	Yes	No	Yes	No	No	No	One year	No	Mailed check	56.95	1889.50	No
2	Male	No	No	No	2	Yes	No	DSL	Yes	Yes	No	No	No	No	Month-to-month	Yes	Mailed check	53.85	108.15	Yes
3	Male	No	No	No	45	No	No phone service	DSL	Yes	No	Yes	Yes	No	No	One year	No	Bank transfer (automatic)	42.30	1840.75	No
4	Female	No	No	No	2	Yes	No	Fiber optic	No	No	No	No	No	No	Month-to-month	Yes	Electronic check	70.70	151.65	Yes

In [25]:

def encode_data(dataframe):
    if dataframe.dtype == "object":
        dataframe = LabelEncoder().fit_transform(dataframe)
    return dataframe

data = data.apply(lambda x: encode_data(x))
data.head()

Out[25]:

	gender	Partner	tenure	PhoneService	MultipleLines	InternetService	OnlineSecurity	OnlineBackup	DeviceProtection	TechSupport	Contract	PaperlessBilling	PaymentMethod	MonthlyCharges	TotalCharges	Churn
0	0	1	1	0	1	0	0	2	0	0	0	1	2	29.85	29.85	0
1	1	0	34	1	0	0	2	0	2	0	1	0	3	56.95	1889.50	0
2	1	0	2	1	0	0	2	2	0	0	0	1	3	53.85	108.15	1
3	1	0	45	0	1	0	2	0	2	2	1	0	0	42.30	1840.75	0
4	0	0	2	1	0	1	0	0	0	0	0	1	2	70.70	151.65	1

In [26]:

X = data.drop(columns = "Churn")
y = data["Churn"].values

In [27]:

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.3, random_state = 42, stratify =y)

In [28]:

def distplot(feature, frame, color='r'):
    plt.figure(figsize=(8,3))
    plt.title("Distribution for {}".format(feature))
    ax = sns.distplot(frame[feature], color= color)

In [29]:

col =  ["tenure", 'MonthlyCharges', 'TotalCharges']
for features in col :distplot(features, data)

In [30]:

data_std = pd.DataFrame(StandardScaler().fit_transform(data[col]).astype('float64'), columns = col)
for feat in col: distplot(feat, data_std, color='c')

In [31]:

data.columns

Out[31]:

Index(['gender', 'SeniorCitizen', 'Partner', 'Dependents', 'tenure',
       'PhoneService', 'MultipleLines', 'InternetService', 'OnlineSecurity',
       'OnlineBackup', 'DeviceProtection', 'TechSupport', 'StreamingTV',
       'StreamingMovies', 'Contract', 'PaperlessBilling', 'PaymentMethod',
       'MonthlyCharges', 'TotalCharges', 'Churn'],
      dtype='object')

In [32]:

for i in data.columns:
    print(i, ": ", data[i].unique())

gender :  [0 1]
SeniorCitizen :  [0 1]
Partner :  [1 0]
Dependents :  [0 1]
tenure :  [ 1 34  2 45  8 22 10 28 62 13 16 58 49 25 69 52 71 21 12 30 47 72 17 27
  5 46 11 70 63 43 15 60 18 66  9  3 31 50 64 56  7 42 35 48 29 65 38 68
 32 55 37 36 41  6  4 33 67 23 57 61 14 20 53 40 59 24 44 19 54 51 26 39]
PhoneService :  [0 1]
MultipleLines :  [1 0 2]
InternetService :  [0 1 2]
OnlineSecurity :  [0 2 1]
OnlineBackup :  [2 0 1]
DeviceProtection :  [0 2 1]
TechSupport :  [0 2 1]
StreamingTV :  [0 2 1]
StreamingMovies :  [0 2 1]
Contract :  [0 1 2]
PaperlessBilling :  [1 0]
PaymentMethod :  [2 3 0 1]
MonthlyCharges :  [29.85 56.95 53.85 ... 63.1  44.2  78.7 ]
TotalCharges :  [  29.85 1889.5   108.15 ...  346.45  306.6  6844.5 ]
Churn :  [0 1]

In [33]:

# Divide the columns into 3 categories, one  for standardisation, one for label encoding and one for one hot encoding

cat_cols_ohe =['PaymentMethod', 'Contract', 'InternetService'] # those that need one-hot encoding
cat_cols_le = list(set(X_train.columns)- set(col) - set(cat_cols_ohe)) #those that need label encoding

print(cat_cols_le)

['PhoneService', 'StreamingTV', 'OnlineSecurity', 'TechSupport', 'SeniorCitizen', 'Dependents', 'Partner', 'MultipleLines', 'DeviceProtection', 'gender', 'StreamingMovies', 'OnlineBackup', 'PaperlessBilling']

In [34]:

scaler = StandardScaler()
X_train[col] = StandardScaler().fit_transform(X_train[col])
X_test[col] = StandardScaler().fit_transform(X_test[col])

In [35]:

models = []

models.append(('Logistic Regression', LogisticRegression(solver='liblinear', random_state = 42, class_weight='balanced')))
models.append(('SVC', SVC(kernel = 'linear', random_state = 42)))
models.append(('KNN', KNeighborsClassifier(n_neighbors = 5, metric = 'minkowski', p = 2)))
models.append(('Gaussian NB', GaussianNB()))
models.append(('Decision Tree Classifier', DecisionTreeClassifier(criterion = 'entropy', random_state = 42)))

In [36]:

acc_results =[]
auc_results =[]
names = []

result_col = ["Algorithm", "ROC AUC Mean", "ROC AUC STD", "Accuracy Mean", "Accuracy STD"]
model_results = pd.DataFrame(columns = result_col)

i=0
# K- fold cross validation

for name, model in models:
    names.append(name)
    kfold = model_selection.KFold(n_splits=10)
    
    cv_acc_results = model_selection.cross_val_score(model, X_train, y_train, 
                    cv = kfold, scoring="accuracy")
    cv_auc_results = model_selection.cross_val_score(model, X_train, y_train,
                    cv = kfold, scoring="roc_auc")
    acc_results.append(cv_acc_results)
    auc_results.append(cv_auc_results)
    
    model_results.loc[i] = [name, 
                           round(cv_auc_results.mean()*100,2),
                           round(cv_auc_results.std()*100,2),
                           round(cv_acc_results.mean()*100,2),
                           round(cv_acc_results.std()*100,2)]
    i+=1

model_results.sort_values(by = ['ROC AUC Mean'], ascending=False)

Out[36]:

	Algorithm	ROC AUC Mean	ROC AUC STD	Accuracy Mean	Accuracy STD
0	Logistic Regression	84.69	2.15	75.01	2.14
1	SVC	83.71	1.89	79.81	1.86
3	Gaussian NB	82.74	2.12	75.70	2.20
2	KNN	76.32	2.30	75.95	2.26
4	Decision Tree Classifier	66.95	2.54	74.16	2.27

2nd Iteration

In [37]:

#evaluation of results
def model_evaluation(y_test, y_pred, model_name):
    acc = accuracy_score(y_test, y_pred)
    prec = precision_score(y_test, y_pred)
    rec = recall_score(y_test, y_pred)
    f1 = f1_score(y_test, y_pred)
    f2 = fbeta_score(y_test, y_pred, beta = 2.0)

    results = pd.DataFrame([[model_name, acc, prec, rec, f1, f2]], 
                       columns = ["Model", "Accuracy", "Precision", "Recall",
                                 "F1 SCore", "F2 Score"])
    results = results.sort_values(["Precision", "Recall", "F2 Score"], ascending = False)
    return results

In [38]:

# Logistic regression
classifier = LogisticRegression(random_state=42)
classifier.fit(X_train, y_train)
y_pred = classifier.predict(X_test)


#SVC
classifier2 = SVC(kernel = 'linear', random_state = 42)
classifier2.fit(X_train, y_train)
y_pred2 = classifier2.predict(X_test)

#KNN
classifier3 = KNeighborsClassifier(n_neighbors=22, metric="minkowski", p=2)
classifier3.fit(X_train, y_train)
y_pred3 = classifier3.predict(X_test)

#Naive Bayes
classifier5 = GaussianNB()
classifier5.fit(X_train, y_train)
y_pred5 = classifier5.predict(X_test)

#Decision tree
classifier6 = DecisionTreeClassifier(criterion="entropy", random_state=42)
classifier6.fit(X_train, y_train)
y_pred6 = classifier6.predict(X_test)

In [39]:

lr = model_evaluation(y_test, y_pred, "Logistic Regression")
svm = model_evaluation(y_test, y_pred2, "SVM (Linear)")
knn = model_evaluation(y_test, y_pred3, "K-Nearest Neighbours")
nb = model_evaluation(y_test, y_pred5, "Naive Bayes")
dt = model_evaluation(y_test, y_pred6, "Decision Tree")

In [40]:

eval_ =lr.append(svm).append(knn).append(nb).append(dt).sort_values(["Precision", 
"Recall", "F2 Score"], ascending = False).reset_index().drop(columns = "index")
eval_

Out[40]:

	Model	Accuracy	Precision	Recall	F1 SCore	F2 Score
0	Logistic Regression	0.801422	0.644898	0.563280	0.601332	0.577908
1	SVM (Linear)	0.794313	0.634249	0.534759	0.580271	0.552079
2	K-Nearest Neighbours	0.769194	0.579060	0.483066	0.526725	0.499631
3	Naive Bayes	0.740284	0.508075	0.729055	0.598829	0.670712
4	Decision Tree	0.733175	0.498239	0.504456	0.501329	0.503201

In [41]:

predictions = [y_pred, y_pred2 , y_pred3,y_pred5, y_pred6]

for i, j in zip(predictions, eval_.Model.values):
    plt.figure(figsize=(4,3))
    sns.heatmap(confusion_matrix(y_test, i),
                annot=True,fmt = "d",linecolor="k",linewidths=3)
    
    plt.title(j,fontsize=14)
    plt.show()

In [42]:

def ROC_curve(classifier_, name, y_pred_):
    classifier_.fit(X_train, y_train) 
    probs = classifier_.predict_proba(X_test) 
    probs = probs[:, 1] 
    classifier_roc_auc = roc_auc_score(y_test, probs )
    rf_fpr, rf_tpr, rf_thresholds = roc_curve(y_test, classifier_.predict_proba(X_test)[:,1])
    plt.figure(figsize=(14, 6))

    label_ = name + '(area = %0.4f)' % classifier_roc_auc
    # Plot Adaboost ROC
    plt.plot(rf_fpr, rf_tpr, 
    label=label_)
    # Plot Base Rate ROC
    plt.plot([0,1], [0,1],label='Base Rate' 'k--')
    plt.xlim([0.0, 1.0])
    plt.ylim([0.0, 1.05])
    plt.ylabel('True Positive Rate \n',horizontalalignment="center",
    fontstyle = "normal", fontsize = "medium", 
    fontfamily = "sans-serif")

    plt.xlabel('\nFalse Positive Rate \n',horizontalalignment="center",
    fontstyle = "normal", fontsize = "medium", 
    fontfamily = "sans-serif")

    plt.title('ROC Graph \n',horizontalalignment="center", 
    fontstyle = "normal", fontsize = "22", 
    fontfamily = "sans-serif")

    plt.legend(loc="lower right", fontsize = "medium")
    plt.xticks(rotation=0, horizontalalignment="center")
    plt.yticks(rotation=0, horizontalalignment="right")
    plt.show()
    
  

In [43]:

preds = [y_pred, y_pred3, y_pred5, y_pred6]
classifiers = [classifier , classifier3, classifier5, classifier6]
model_names_ = ["Logistic Regression", "K-Nearest Neighbours","Naive Bayes",
               "Decision Tree"]

for i, j, k in zip(classifiers, model_names_, predictions):
    ROC_curve(i, j, k) 