基于神经网络的电商消费者复购行为预测

Prediction of Repurchase Behavior of E-Commerce Consumers Based on Neural Network

摘要: 随着互联网零售行业持续高涨，越来越多的消费者选择在网络购物，尤其是中国消费者。利用消费者在网络留下的行为数据进行重复购买行为的预测对企业实现精准营销有着重要的意义。本文采用电子商务平台上的历史行为数据对消费者复购行为进行预测有助于提升用户体验和营销效果。提出一种基于神经网络的消费者复购行为预测模型，实现用户属性、商品属性及用户行为特征的自动抽取与选择，并以此对消费者复购行为进行预测。在阿里巴巴移动电商平台数据集的实验结果表明，基于神经网络的预测模型F1值比基准模型平均提升了7%~11%。

Abstract: As the Internet retail industry continues to rise, more and more consumers choose to shop online, especially Chinese consumers. Using consumer behavior data left on the Internet to predict repeat purchase behavior is of great significance for companies to achieve precision marketing. This article uses historical behavior data on the e-commerce platform to predict consumer repurchase behavior, which will help improve user’s experience and marketing effects. A neural network-based consumer repurchase behavior prediction model is proposed to realize the automatic extraction and selection of user’s attributes, commodity attributes and user’s behavior characteristics, and to predict consumer repurchase behaviors. Experimental results on the data set of Alibaba’s mobile e-commerce platform show that the F1 value of the neural network-based prediction model is improved by an average of 7%~11% compared with the benchmark model.