Title: Prediction method of consumer repeat purchase behaviour in e-commerce environment

Authors: Miao Sun

Addresses: School of Management, Changchun University of Architecture and Civil Engineering, Changchun 130604, China

Abstract: Aiming at the problems of poor prediction accuracy and large prediction time cost of traditional repeat purchase behaviour prediction methods, a new prediction method of consumer repeat purchase behaviour in e-commerce environment is designed. Firstly, the factors affecting consumers' repeat purchase are determined according to the subjective and objective factors, and the characteristics of consumers' repeat purchase behaviour are extracted according to different types of characteristic data. Secondly, K-means algorithm is used to determine the clustering centre point of all feature data, and inconsistent feature data are removed according to the set standard conditions. Finally, a prediction model of consumer repeat purchase behaviour in e-commerce environment is constructed. The experimental results show that this method can quickly and accurately obtain the prediction results of e-commerce repeat purchase behaviour, and the prediction accuracy is always higher than 90%.

Keywords: e-commerce; electronic commerce; consumer behaviour; repeat purchase; behaviour prediction; K-means algorithm.

DOI: 10.1504/IJWBC.2023.134868

International Journal of Web Based Communities, 2023 Vol.19 No.4, pp.267 - 278

Received: 15 Dec 2021
Accepted: 06 May 2022
Published online: 15 Nov 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article