Customer shopping pattern prediction: A recurrent neural network approach

Customer relationship management is a popular and strategic topic in marketing and quality of service. The availability of big transactions data as well as computing systems have provided a great opportunity to model and predict customer behaviour. However, there is a lack of modern modelling and analytical methods to perform analysis on such data. Deep learning techniques can assist marketing decision makers to provide more reliable and practical marketing strategic plans. In this paper, we propose a customer behaviour prediction model using recurrent neural networks (RNNs) based on the client loyalty number (CLN), recency, frequency, and monetary (RFM) variables. The experiment results show that RNNs can predict RFM values of customers efficiently. This model can be later used in recommender systems for exclusive promotional offers and loyalty programs management.