Predict Churn for a Telecom company using Logistic Regression

Predict Churn for a Telecom company using Logistic Regression

Machine Learning Project in R- Predict the customer churn of telecom sector and find out the key drivers that lead to churn. Learn how the logistic regression model using R can be used to identify the customer churn in telecom dataset.
explanation image


Each project comes with 2-5 hours of micro-videos explaining the solution.

ipython image

Code & Dataset

Get access to 50+ solved projects with iPython notebooks and datasets.

project experience

Project Experience

Add project experience to your Linkedin/Github profiles.

Customer Love

Read All Reviews
profile image

Hiren Ahir linkedin profile url

Microsoft Azure SQL Sever Developer, BI Developer

I'm a Graduate student and came into the job market and found a university degree wasn't sufficient to get a good paying job. I aimed at hottest technology in the market Big Data but the word BigData... Read More

profile image

Arvind Sodhi linkedin profile url

VP - Data Architect, CDO at Deutsche Bank

I have extensive experience in data management and data processing. Over the past few years I saw the data management technology transition into the Big Data ecosystem and I needed to follow suit. I... Read More

What will you learn

Understand the Problem Statement
Perform basic EDA to get familiarize with the data
Take care of any missing values or datatype issues in the data
Perform Univariate analysis for both numeric and categorical variables
Perform Bi-variate analysis to identify redundant variables
Plot Probability Distribution Function (PDF) of each predictor the target variable
Create new features that might add value to the model
Define a function for each set of code that might need to be repeated again
Prepare the data for modelling
Create models using multiple algorithms
Perform Hyper-parameter tuning to get the best parameters
Compare the performance of different models using multiple metrics
Apply feature reduction methods to get a light model while maintaining performance
Check whether the top model features are as per your business understanding/intuition
Save the model and creating API to implement the model in production

Project Description

Every company wants to increase its revenue and profitability. To do that, while they acquire new customers, they also want to make sure that the existing ones stay with them for a long term. Also, its strategically important to know beforehand whether a set of customers are planning to stop using their services (especially recurring ones like internet, cable, phone etc.). To do that, every company or business creates and tracks customer metrics which are then used to predict their likelihood of churn.

Customer Churn for a company occurs when a customer decides to stop using the services of that company. In this project, we will be using the customer data of a telecom sector company based in the US to predict the probability of churn for each of the customer. We will look at the standard practices that are followed in the industry to solve these problems and also go beyond just those techniques. We have chosen the telecom company data for the churn problem as it is a major area of concern for companies in that sector.

Once we have built a model, the churn model output can also be used as a warning indicator that some customers are likely to churn. The key drivers that are making the customer more likely to churn can be alleviated and ensure that the customers are actually retained.

Similar Projects

Data Science Project in R -Build a machine learning algorithm to predict the future sale prices of homes.

In this machine learning project, you will build predictive models to identify wine preferences of people using physiochemical properties of wines and help restaurants recommend the right quality of wine to a customer.

In this machine learning project, you will uncover the predictive value in an uncertain world by using various artificial intelligence, machine learning, advanced regression and feature transformation techniques.

Curriculum For This Mini Project

Problem Statement
Exploratory Data Analysis - 1
Exploratory Data Analysis - 2
Converting Categorical To Numeric Variables
Univariate Analysis - Non Continuous Variables
Univariate Analysis - Continuous Variables
Bivariate Analysis
Feature Creation
Probability Distribution Function
Preparing Data For Modelling
Performance Tuning - 1
Performance Tuning - 2
Performance Tuning - 3
Recursive Feature Elimination - 1
Recursive Feature Elimination - 2
Key Drivers Of Churn-1
Model Implementation - 1
Model Implementation - 2