TELECOM CUSTOMER CHURN PREDICTION 📈 Did you know that attracting a new customer costs…
- User
TELECOM CUSTOMER CHURN PREDICTION 📈 Did you know that attracting a new customer costs five times as much as keeping an existing one?
linkcode Content Introduction What is Customer Churn? How can customer churn be reducded? Objectives Loading libraries and data Undertanding the data Visualize missing values Data Manipulation Data Visualization Data Preprocessing Standardizing numeric attributes Machine Learning Model Evaluations and Predictions KNN SVC Random Forest Logistic Regression Decision Tree Classifier AdaBoost Classifier Gradient Boosting Classifier Voting Classifier Q- and this my Code here
import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns
path="data/raw/data.csv" df=pd.read_csv(path)
Next
Dataset Information
The dataset includes the following key details:
-
Customers who left within the last month:
- Column:
Churn
- Column:
-
Services that each customer has signed up for:
- Phone
- Multiple lines
- Internet
- Online security
- Online backup
- Device protection
- Tech support
- Streaming TV and movies
-
Customer account information:
- How long they’ve been a customer
- Contract type
- Payment method
- Paperless billing
- Monthly charges
- Total charges
-
Demographic information about customers:
- Gender
- Age range
- Whether they have partners
- Whether they have dependents
Next
Df.shape ((7043, 21)
Next
df.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 Here’s a table that classifies the features of the DataFrame into numerical and categorical:Feature Type Category customerIDObject Categorical genderObject Categorical SeniorCitizenInt64 Numerical PartnerObject Categorical DependentsObject Categorical tenureInt64 Numerical PhoneServiceObject Categorical MultipleLinesObject Categorical InternetServiceObject Categorical OnlineSecurityObject Categorical OnlineBackupObject Categorical DeviceProtectionObject Categorical TechSupportObject Categorical StreamingTVObject Categorical StreamingMoviesObject Categorical ContractObject Categorical PaperlessBillingObject Categorical PaymentMethodObject Categorical MonthlyChargesFloat64 Numerical TotalChargesObject Categorical (should be numerical, but likely has non-numeric values) ChurnObject Categorical Notes:
- Numerical features: These are the ones that contain numeric data types (
int64,float64). - Categorical features: These are
objecttypes that generally represent categories. - The
TotalChargescolumn is an object, but it might need conversion to numeric (after cleaning) since it's likely a numeric field stored as a string due to potential non-numeric values. in percentage how % mnay categoricla vs nurmerical here
-
- Assistant
Certainly! Let's break down the distribution of categorical versus numerical features in your dataset.
Dataset Overview
- Total Features (Columns): 21
Feature Classification
Feature Type Category customerIDObject Categorical genderObject Categorical SeniorCitizenInt64 Numerical PartnerObject Categorical DependentsObject Categorical tenureInt64 Numerical PhoneServiceObject Categorical MultipleLinesObject Categorical InternetServiceObject Categorical OnlineSecurityObject Categorical OnlineBackupObject Categorical DeviceProtectionObject Categorical TechSupportObject Categorical StreamingTVObject Categorical StreamingMoviesObject Categorical ContractObject Categorical PaperlessBillingObject Categorical PaymentMethodObject Categorical MonthlyChargesFloat64 Numerical TotalChargesObject Categorical (should be Numerical) ChurnObject Categorical Percentage Breakdown
-
Numerical Features:
SeniorCitizentenureMonthlyCharges
Count: 3
-
Categorical Features:
customerID,gender,Partner,Dependents,PhoneService,MultipleLines,InternetService,OnlineSecurity,OnlineBackup,DeviceProtection,TechSupport,StreamingTV,StreamingMovies,Contract,PaperlessBilling,PaymentMethod,TotalCharges,Churn
Count: 18
Calculations
-
Percentage of Categorical Features:
[ \left( \frac{18}{21} \right) \times 100 \approx 85.71% ]
-
Percentage of Numerical Features:
[ \left( \frac{3}{21} \right) \times 100 \approx 14.29% ]
Summary
- 85.71% of the features are Categorical.
- 14.29% of the features are Numerical.
Recommendation
-
TotalCharges: Although currently classified as Categorical due to itsobjectdata type, this feature is inherently numerical. It's advisable to clean and convert this column to a numeric type (e.g.,float64) to leverage its quantitative information effectively in your machine learning models.# Convert 'TotalCharges' to numeric, coercing errors to NaN df['TotalCharges'] = pd.to_numeric(df['TotalCharges'], errors='coerce') # Handle missing values if any after conversion df['TotalCharges'].fillna(df['TotalCharges'].median(), inplace=True)After conversion, you can reclassify
TotalChargesas a Numerical feature, adjusting the percentages accordingly.
Feel free to ask if you need further assistance with data preprocessing or model building!