北京邮电大学学报(社科版) ›› 2020, Vol. 22 ›› Issue (3): 7-14.doi: 10.19722/j.cnki.1008-7729.2020.0056

• 电子商务 • 上一篇    下一篇

基于Pareto/NBD的用户价值预测模型研究

  

  1. 1.北京邮电大学 经济管理学院,北京 100876;2.中国科学院科技战略咨询研究院,北京100190;
    3.滴滴(中国)科技有限公司,北京100193
  • 收稿日期:2020-03-12 出版日期:2020-06-30
  • 基金资助:
    国家自然科学基金资助项目(71974182)

Prediction Model of Consumer Value Based on Pareto/NBD

  1. 1. School of Economics and Management, Beijing University of Posts and Telecommunications, Beijing
    100876, China;2. Institutes of Science and Development, Chinese Academy of Sciences,
    Beijing 100190, China;3. Didi (China) Technology Co., Ltd, Beijing 100193, China
  • Received:2020-03-12 Online:2020-06-30

摘要: 为提高用户价值预测的准确率,在已有Pareto/NBD模型基础上,结合有监督的机器学习算法提出改进模型,即使用最近多个子时间段的购买行为数据来共同预测未来购买频次,充分利用用户历史消费信息。最后,使用滴滴(中国)科技有限公司用户消费数据对改进的模型进行验证。结果表明,改进后的Pareto/NBD模型显著提高了预测准确率。

关键词: 用户生命周期价值, Pareto/NBD, 有监督学习

Abstract: To improve the prediction accuracy of consumer value,based on the existing Pareto/NBD model and the supervised machine learning algorithm, an improved model is proposed, that is, the purchasing behavior data of recent sub-periods are used to predict the future purchasing frequency, which makes full use of users historical consumption information. The improved model was verified by using DiDis user consumption data, and the results show that the proposed model can improve the prediction accuracy significantly.

Key words: customer lifecycle value, Pareto/NBD, supervised learning model

中图分类号: