Churn prediction analysis of telecom customers using svm, random forest and logistic regression models using orange data mining tools

Abstract

This research aims to apply classification algorithms to telecommunication customer churn data using Orange Data Mining. The methods used include Support Vector Machine (SVM), Random Forest and Logistic Regression. The dataset used is secondary data, the dataset is downloaded from the kaggle website with a total of 7,043 customer data and 21 variables that will be used to predict telecommunication churn and in this study Exploratory Data Analysis (EDA) was conducted to understand the characteristics of the data and identify patterns and trends that can be used to improve the performance of classification algorithms. The results of EDA show that telecommunication customer churn data has several characteristics, namely unbalanced churn data, with the number of customers who churn less than customers who do not churn. With the results of the accuracy value, namely Random Forest 76% followed by Logistic Regression 79% and SVM 74%. The best accuracy is obtained by Logistic Regression with an accuracy value of 79%. These results show that logistic regression has a better ability to classify telecommunication customer churn data compared to other classification algorithms, this research shows that Orange Data Mining can be used to classify telecommunication customer churn data.