Comparative Analysis of Accuracy and Prediction of Customer Loyalty in the Telecom Industry using Novel Diverse Algorithm

Abstract

The goal of this project is to see how well a different algorithm can predict customer loyalty in the telecom industry. The customer information dataset used to train and test the proposed prediction model includes 7043 customers with 21 different traits. It is done by adapting Random Forest (RF), Logistic Regression (LR), K-Nearest Neighbors (KNN), and Support Vector Machine (SVM) algorithms to Figure out how many people will leave. Logistic Regression classifier is (80%), KNN classifier is (78%), and SVM classifier is (90%). (75 percent). With a significance value (p0.05), there is a big difference between the study groups. People who did this experiment say it’s clear that the Logistic Regression classifier does better than Random Forest, KNN, or SVM in terms of both precision and accuracy.