Customer Churn Prediction Modelling Based on Behavioural Patterns Analysis using Deep Learning

Abstract

Customer churn refers to when a customer ceases their relationship with a company. A churn rate, used to estimate growth, is now considered as important a metric as financial profit. With growing competition in the market, companies are desperate to keep the churn rate as low as possible. Thus, churn prediction has gained critical importance, not just for existing customers, but also for predicting trends of future customers. This paper demonstrates prediction of churn on a Telco dataset using a Deep Learning Approach. A multilayered Neural Network was designed to build a non-linear classification model. The churn prediction model works on customer features, support features, usage features and contextual features. The possibility of churn as well as the determining factors are predicted. The trained model then applies the final weights on these features and predict the possibility of churn for that customer. An accuracy of 80.03% was achieved. Since the model also provides the churn factors, it can be used by companies to analyze the reasons for these factors and take steps to eliminate them.

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 — 59 citations exist upstream.

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