Comparative Methods for Personalized Customer Churn Prediction with Sequential Data

• Published in: 2022 IEEE International Conference on Big Data and Smart Computing (BigComp) pp. 222 - 225
• Main Authors: Bayrak, Ahmet Tugrul, Yuceturk, Guven, Bahadir, Musa Berat, Yalcinkaya, Sare Melek, Demirdag, Melike, Sayan, Ismail Utku
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.01.2022
• Subjects: churn analysis; Computational modeling; Conferences; convolutional neural network; Data models; deep learning; Industries; Logic gates; Predictive models; recurrent neural network; Recurrent neural networks; sequential data
• ISSN: 2375-9356
• DOI: 10.1109/BigComp54360.2022.00050

Competition in the business environment has been increasing with the developed technology. There are various alternative service providers for customers. Those providers would like to retain their current customers. To reach that goal, churn prediction is a convenient method. However, predicting churning customers is not a trivial task. It is more demanding for some sectors like fast-food, since there can be numerous reasons when a customer stops having services. In this study, the challenging situation of customer churn in the fast-food industry is analyzed, and a fast-food chain's data is used. The data is formed sequentially accordingly with customers' personal churn periods. Several recurrent neural network models such as gated recurrent units and long short-term memory are built using the sequential data to predict the churn stages of customers, and they are compared with the other standard classification methods. Apart from recurrent neural network models, a hybrid model consisting of a convolutional network and long short-term memory is applied.