A New 360° Framework to Predict Customer Lifetime Value for Multi-Category E-Commerce Companies Using a Multi-Output Deep Neural Network and Explainable Artificial Intelligence

Online purchasing has developed rapidly in recent years due to its efficiency, convenience, low cost, and product variety. This increased the number of online multi-category e-commerce retailers that sell a variety categories. Due growth players, each company needs optimize own business strategy order compete. Customer lifetime value (CLV) is common metric usually consider for competition because it helps determine most valuable customers retailers. However, this paper, we introduce two additional novel factors addition CLV which will bring highest revenue future: distinct category (DPC) trend amount spent (TAS). Then, propose new framework. We utilized, first time relevant literature, multi-output deep neural network (DNN) model test our proposed framework while forecasting CLV, DPC, TAS together. To make outcome applicable real life, constructed customer clusters allow management companies segment end-users based on three variables. compared (constructed with multiple outputs: TAS) against baseline single-output combined effect model. In addition, also Decision Tree (DT) Random Forest (RF) algorithms same dataset. The results indicate DNN outperforms model, DT, RF across all assessment measures, proving more suitable retailers’ usage. Furthermore, Shapley values derived through explainable artificial intelligence method are used interpret decisions DNN. practice demonstrates inputs contribute outcomes (a significant novelty interpreting CLV).