Machine Learning-based Prediction Mechanism of Repeated Purchase Behavior of E-commerce Customers

A machine learning-based prediction model was developed to understand the repeated purchase behavior of e-commerce customers. A data mining process and the logistic regression algorithm were used in the model. Data preprocessing such as data cleaning and feature engineering was conducted for training the model. The model was evaluated in the experiment for accuracy, precision, recall, and area under the receiver operating characteristics curve to ensure comprehensive assessment. A fusion approach was used to combine multiple machine-learning models to enhance prediction accuracy and robustness. Bagging, boosting, stacking, and voting were employed to leverage the strengths of different models for the robust prediction of the model. The model allows e-commerce companies to effectively predict and enhance customer retention.