259

📝 Dataset Description This dataset contains information about customers of a telecommunications company, including their demographic details, account information, service subscriptions, and churn status. It is a modified version of the popular Telco Churn dataset, curated for exploratory data analysis, machine learning model development, and churn prediction tasks.

The dataset includes simulated missing values in some columns to reflect real-world data issues and support preprocessing and imputation tasks. This makes it especially useful for demonstrating data cleaning techniques and evaluating model robustness.

📂 Files Included telco_data_modified.csv: The main dataset with 21 columns and 7043 rows (some missing values are intentionally inserted).

📌 Features Column Name Description customerID Unique identifier for each customer gender Customer gender: Male/Female SeniorCitizen Indicates if the customer is a senior citizen (0 = No, 1 = Yes) Partner Whether the customer has a partner Dependents Whether the customer has dependents tenure Number of months the customer has stayed with the company PhoneService Whether the customer has phone service MultipleLines Whether the customer has multiple lines InternetService Customer's internet service provider (DSL, Fiber optic, No) OnlineSecurity Whether the customer has online security OnlineBackup Whether the customer has online backup DeviceProtection Whether the customer has device protection TechSupport Whether the customer has tech support StreamingTV Whether the customer has streaming TV StreamingMovies Whether the customer has streaming movies Contract Type of contract: Month-to-month, One year, Two year PaperlessBilling Whether the customer uses paperless billing PaymentMethod Payment method: (e.g., Electronic check, Mailed check, etc.) MonthlyCharges Monthly charges TotalCharges Total charges to date Churn Whether the customer has left the company (Yes/No)

🔍 Use Cases Binary classification: Predict customer churn

Data preprocessing and imputation exercises

Feature engineering and importance analysis

Customer segmentation and churn modeling

⚠️ Notes Missing values were intentionally inserted in the dataset to help simulate real-world conditions.

Some preprocessing may be required before modeling (e.g., converting categorical to numerical data, handling TotalCharges as numeric).

🏷️ Tags

telecom #churn #classification #customer-analytics #data-cleaning #feature-engineering

🙏 Acknowledgements This dataset is based on the original Telco Customer Churn dataset (initially provided by IBM). The current version has been modified for academic and practical exercises.

Customer churn prediction dataset of a fictional telecommunication company made by IBM Sample Datasets. Context Predict behavior to retain customers. You can analyze all relevant customer data and develop focused customer retention programs. Content Each row represents a customer, each column contains customer’s attributes described on the column metadata. The data set includes information about:

Customers who left within the last month: the column is called Churn Services that each customer… See the full description on the dataset page: https://huggingface.co/datasets/scikit-learn/churn-prediction.

60,000+ Real Anonymized Customer Usage Data for Churn Prediction!

Dataset Information

• Dataset Name: Real World Customer Churn Dataset in Telco Domain
• Snapshot Period: January 1, 2023, to March 31, 2023
• Source: One of the Largest Telco Companies in Sri Lanka
• Data Anonymization: The Dataset is Anonymized to Protect Customer Privacy.

Overview

The "Real World Customer Churn Dataset in Telco Domain" is a comprehensive collection of anonymized data that provides insights into customer behavior and churn prediction within the telecommunications industry.

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F6361330%2F860271e0362e6c10503889f289201402%2FCustomer-churn.jpg?generation=1698182677600097&alt=media" alt="Dataset Image">

Usage Categories

The dataset contains data on over 60,000 customers across more than 10+ distinct usage categories. Some of the key usage categories include:

• usage_app_youtube_daily: YouTube Traffic in MBs.
• usage_app_facebook_daily: Facebook Traffic in MBs.
• usage_app_tiktok_daily: TikTok Traffic in MBs.
• usage_app_whatsapp_daily: WhatsApp Traffic in MBs.
• usage_app_helakuru_daily: Helakuru App traffic in MBs.
• usage_voice_o2o_outgoing: Outgoing call volume in minutes between the same operator.
• usage_voice_o2op_outgoing: Outgoing call volume in minutes between operator and other operators.
• usage_voice_o2o_incoming: Incoming call volume in minutes between the same operator.
• usage_voice_op2o_incoming: Incoming call volume in minutes between other operator to operator.
• usage_pack_data: Spend in LKR for data package purchasing.
• usage_pack_vas: Spend in LKR for value-added service rentals or usage.

Dataset Files

The dataset consists of the following key files:

1. main.csv: An aggregated dataset that compiles usage data from all usage categories, providing a holistic view of customer behavior.
2. raw_dump folder: The raw data export, preserving the original source data for detailed exploration.
3. test and train folders: These folders contain customer IDs and corresponding Churn Labels, facilitating model training and testing.
4. usage_profiles folder: It comprises broken-down data frames for each customer under specific usage categories, allowing in-depth analysis of individual customer behavior within various usage categories.

Potential Use Cases

The "Real World Customer Churn Dataset in Telco Domain" offers a range of potential use cases, including:

• Customer Churn Prediction: Leveraging customer usage patterns to predict and reduce churn.
• Targeted Marketing: Designing customized marketing campaigns based on customer preferences.
• Service Quality Enhancement: Identifying areas for service improvement, such as network quality.
• Revenue Optimization: Maximizing revenue through the analysis of data package spending and value-added service usage.

Dataset Importance

This dataset's real-world aspect is of significant importance. It reflects actual customer interactions with a major telecommunications company in Sri Lanka, offering insights that can be directly applied to real-world scenarios. The dataset is sourced from one of the largest telco companies in the country, adding credibility and relevance to the insights it provides.

Understanding customer churn and usage behavior is pivotal for the telecommunications industry, and this dataset empowers researchers, data scientists, and businesses to gain deeper insights into these aspects.

Disclaimer

The dataset is anonymized to protect customer privacy, and all data used is in compliance with privacy regulations and agreements. Users are encouraged to explore and contribute to the "Real World Customer Churn Dataset in Telco Domain."

Thank you for your valuable contributions to this dataset.

Context and Methodology

The dataset originates from the research domain of Customer Churn Prediction in the Telecom Industry. It was created as part of the project "Data-Driven Churn Prediction: ML Solutions for the Telecom Industry," completed within the Data Stewardship course (Master programme Data Science, TU Wien).

The primary purpose of this dataset is to support machine learning model development for predicting customer churn based on customer demographics, service usage, and account information.
The dataset enables the training, testing, and evaluation of classification algorithms, allowing researchers and practitioners to explore techniques for customer retention optimization.

The dataset was originally obtained from the IBM Accelerator Catalog and adapted for academic use. It was uploaded to TU Wien’s DBRepo test system and accessed via SQLAlchemy connections to the MariaDB environment.

Technical Details

The dataset has a tabular structure and was initially stored in CSV format. It contains:

• Rows: 7,043 customer records

• Columns: 21 features including customer attributes (gender, senior citizen status, partner status), account information (tenure, contract type, payment method), service usage (internet service, streaming TV, tech support), and the target variable (Churn: Yes/No).

Naming Convention:

• The table in the database is named telco_customer_churn_data.

Software Requirements:

• To open and work with the dataset, any standard database client or programming language supporting MariaDB connections can be used (e.g., Python etc).

• For machine learning applications, libraries such as pandas, scikit-learn, and joblib are typically used.

Additional Resources:

• Source code for data loading, preprocessing, model training, and evaluation is available at the associated GitHub repository: https://github.com/nazerum/fair-ml-customer-churn

Further Details

When reusing the dataset, users should be aware:

• Licensing: The dataset is shared under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

• Use Case Suitability: The dataset is best suited for classification tasks, particularly binary classification (churn vs. no churn).

• Metadata Standards: Metadata describing the dataset adheres to FAIR principles and is supplemented by CodeMeta and Croissant standards for improved interoperability.

RavenStack is a fictional AI-powered collaboration platform used to simulate a real-world SaaS business. This simulated dataset was created using Python and ChatGPT specifically for people learning data analysis, business intelligence, or data science. It offers a realistic environment to practice SQL joins, cohort analysis, churn modeling, revenue tracking, and support analytics using a multi-table relational structure.

The dataset spans 5 CSV files:

• accounts.csv – customer metadata

• subscriptions.csv – subscription lifecycles and revenue

• feature_usage.csv – daily product interaction logs

• support_tickets.csv – support activity and satisfaction scores

• churn_events.csv – churn dates, reasons, and refund behaviors

Users can explore trial-to-paid conversion, MRR trends, upgrade funnels, feature adoption, support patterns, churn drivers, and reactivation cycles. The dataset supports temporal and cohort analyses, and has built-in edge cases for testing real-world logic.

Analysis of ‘Customer Churn’ provided by Analyst-2 (analyst-2.ai), based on source dataset retrieved from https://www.kaggle.com/hassanamin/customer-churn on 13 November 2021.

--- Dataset description provided by original source is as follows ---

Binary Customer Churn

A marketing agency has many customers that use their service to produce ads for the client/customer websites. They've noticed that they have quite a bit of churn in clients. They basically randomly assign account managers right now, but want you to create a machine learning model that will help predict which customers will churn (stop buying their service) so that they can correctly assign the customers most at risk to churn an account manager. Luckily they have some historical data, can you help them out? Create a classification algorithm that will help classify whether or not a customer churned. Then the company can test this against incoming data for future customers to predict which customers will churn and assign them an account manager.

Content

The data is saved as customer_churn.csv. Here are the fields and their definitions:

Name : Name of the latest contact at Company

Age: Customer Age

Total_Purchase: Total Ads Purchased

Account_Manager: Binary 0=No manager, 1= Account manager assigned

Years: Totaly Years as a customer

Num_sites: Number of websites that use the service.

Onboard_date: Date that the name of the latest contact was onboarded

Location: Client HQ Address

Company: Name of Client Company

Once you've created the model and evaluated it, test out the model on some new data (you can think of this almost like a hold-out set) that your client has provided, saved under new_customers.csv. The client wants to know which customers are most likely to churn given this data (they don't have the label yet).

Acknowledgements

We wouldn't be here without the help of others. If you owe any attributions or thanks, include them here along with any citations of past research.

Inspiration

Your data will be in front of the world's largest data science community. What questions do you want to see answered?

--- Original source retains full ownership of the source dataset ---

"It's necessary to implement an effective customer retention strategy through data analysis. The main goal is to predict the probability of customer churn for the next month, identify key customer profiles, and develop specific recommendations to improve customer retention and satisfaction. This will enable optimizing the customer experience and strengthening their loyalty."

Nowadays

PLEASE UPVOTE THIS DATASET IF THIS HELP YOU... GLAD TO ANY FORKS HERE

BACKGROUND DQLab Telco is a telecommunications company with numerous locations all over the world. In order to ensure that customers are not left behind, DQLab Telco has consistently paid attention to the customer experience since its establishment in 2019.

Even though DQLab Telco is only a little over a year old, many of its customers have already changed their subscriptions to rival companies. By using machine learning, management hopes to lower the number of customers who leave.

After cleaning the data yesterday, it is now time for us to build the best model to forecast customer churn.

TASKS & STEPS Yesterday, we completed "Cleansing Data" as part of project part 1. You are now expected to develop the appropriate model as a data scientist.

You will perform "Machine Learning Modeling" in this assignment using data from the previous month, specifically June 2020.

The actions that must be taken are, 1. Analyze exploratory data first. 2. Carry out pre-processing of the data. 3. Using modeling from machine learning. 4. Picking the Ideal Model.

Dataset

This dataset was created by YiChien_Chong

Contents

Predicting Credit Card Customer Segmentation

Exploring Key Customer Characteristics

By [source]

About this dataset

This dataset contains a wealth of customer information collected from within a consumer credit card portfolio, with the aim of helping analysts predict customer attrition. It includes comprehensive demographic details such as age, gender, marital status and income category, as well as insight into each customer’s relationship with the credit card provider such as the card type, number of months on book and inactive periods. Additionally it holds key data about customers’ spending behavior drawing closer to their churn decision such as total revolving balance, credit limit, average open to buy rate and analyzable metrics like total amount of change from quarter 4 to quarter 1, average utilization ratio and Naive Bayes classifier attrition flag (Card category is combined with contacts count in 12months period alongside dependent count plus education level & months inactive). Faced with this set of useful predicted data points across multiple variables capture up-to-date information that can determine long term account stability or an impending departure therefore offering us an equipped understanding when seeking to manage a portfolio or serve individual customers

More Datasets

For more datasets, click here.

Featured Notebooks

• 🚨 Your notebook can be here! 🚨!

How to use the dataset

This dataset can be used to analyze the key factors that influence customer attrition. Analysts can use this dataset to understand customer demographics, spending patterns, and relationship with the credit card provider to better predict customer attrition.

Research Ideas

• Using the customer demographics, such as gender, marital status, education level and income category to determine which customer demographic is more likely to churn.
• Analyzing the customer’s spending behavior leading up to churning and using this data to better predict the likelihood of a customer of churning in the future.
• Creating a classifier that can predict potential customers who are more susceptible to attrition based on their credit score, credit limit, utilization ratio and other spending behavior metrics over time; this could be used as an early warning system for predicting potential attrition before it happens

Acknowledgements

If you use this dataset in your research, please credit the original authors. Data Source

License

License: CC0 1.0 Universal (CC0 1.0) - Public Domain Dedication No Copyright - You can copy, modify, distribute and perform the work, even for commercial purposes, all without asking permission. See Other Information.

Columns

File: BankChurners.csv | Column name | Description | |:---------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------| | CLIENTNUM | Unique identifier for each customer. (Integer) | | Attrition_Flag | Flag indicating whether or not the customer has churned out. (Boolean) | | Customer_Age | Age of customer. (Integer) | | Gender | Gender of customer. (String) | | Dependent_count | Number of dependents that customer has. (Integer) | | Education_Level ...

Content

Insurance companies around the world operate in a very competitive environment. With various aspects of data collected from millions of customers, it is painstakingly hard to analyse and understand the reason for a customer’s decision to switch to a different insurance provider.

For an industry where customer acquisition and retention are equally important, and the former being a more expensive process, insurance companies rely on data to understand customer behavior to prevent retention. Thus knowing whether a customer is possibly going to switch beforehand gives Insurance companies an opportunity to come up with strategies to prevent it from actually happening.

Given are 16 distinguishing factors that can help in understanding the customer churn, your objective as a data scientist is to build a Machine Learning model that can predict whether the insurance company will lose a customer or not using these factors.

Data Description

The unzipped folder will have the following files.

Train.csv – 33908 observations. Test.csv – 11303 observations. Sample Submission – Sample format for the submission.

Target Variable: labels

Acknowledgements

The Dataset comes from the following link: https://www.machinehack.com/course/insurance-churn-prediction-weekend-hackathon-2/