Prediction of Churning Behavior of Customers in Telecom Sector Using Supervised Learning Techniques

Abstract

Data mining is vast area that co-relates different branches i.e Statistics, Data Base, Machine learning and Artificial intelligence. Numbers of applications are available in different sectors. Customer churn is the behavior when customer no longer wants to keep his relationship with the company. Customer churn management is playing important role in customer management. Nowadays telecommunication companies are focusing on identifying high value and potential churning customers to increase profit and market share. It is understood that making new customer is more expensive rather than retaining existing customer. There is an existing problem that customers leave the company due to unknown reasons. In our research we predict churn behavior of customer by using various data mining techniques. It will eventually help in analyzing customer's behavior and classify whether it is a churning customer or not. In this research, we used online data set available at Kaggle for prediction of Customer churn behavior using different classifiers i.e SVM (Support Vector Machine), Bagging, Stacking, C50/J48, PART, Naïve Bayes, Baysen Net, Adaboost and observe that our model gave 99.8% accuracy level using Bagging Algorithms.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 15 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.