Dynamic Bayesian Network–Based Product Recommendation Considering Consumers' Multistage Shopping Journeys: A Marketing Funnel Perspective.

Wei, Qiang; Mu, Yao; Guo, Xunhua; Jiang, Weijie; Chen, Guoqing

Recommender systems are widely used by platforms/merchants to find the products that are likely to interest consumers. However, existing dynamic methods still face challenges with regard to diverse behaviors, variability in interest shifts, and the identification of psychological dynamics. Premised on the marketing funnel perspective to analyze consumer shopping journeys, this study proposes a novel and effective machine learning approach for product recommendation, namely, multi-stage dynamic Bayesian network (MS-DBN), which models the generative processes of consumers' interactive behaviors with products in light of their stage transitions and interest shifts. In this way, consumers' stage-interest-behavior dynamics can be learnt, especially the variability in interest shifts. This provides managerial implications for practice. MS-DBN demonstrates significant performance advantage with general applicability by extracting the generalizable regularity during shopping journeys, which compensates the diversity and sparsity frequently observed in consumer behaviors. In addition, aided by the identification strategies integrated into the learning process, the latent variables in the model can be detected such that consumers' invisible psychological stages and interests in products can be identified from their observed behaviors, shedding light on the targeted marketing of platforms/merchants and thus enriching the practical value of the approach. Recommender systems are widely used by online merchants to find the products that are likely to interest consumers, but existing dynamic methods still face challenges regarding diverse behaviors, variability in interest shifts, and the identification of psychological dynamics. Premised on the marketing funnel perspective to analyze consumer shopping journeys, this study proposes a novel machine learning approach for product recommendation, namely, multistage dynamic Bayesian network (MS-DBN), to model the generative processes of consumers' interactive behaviors with products in light of stage transitions and interest shifts. This approach features a dynamic Bayesian network model to overcome the problem of diverse behaviors and extract generalizable regularity of consumers' psychological dynamics, two latent layers to depict variability in consumers' interest shifts across multiple stages, and the identification strategies that dynamically detect the invisible stages and interests of consumers. Extensive experiments on large-scale real-world data and comprehensive robustness checks manifest the superior performance of the proposed MS-DBN approach over baseline methods. History: Olivia Liu Sheng, Senior Editor; Huimin Zhao, Associate Editor. Funding: This work was supported by the National Natural Science Foundation of China [Grants 72172070, 72302153, 72293561, and 92246001]. Supplemental Material: The online appendices are available at https://doi.org/10.1287/isre.2020.0277.

CONSUMER behavior; BAYESIAN analysis; LATENT variables; CONSUMERS; MACHINE learning; RECOMMENDER systems

Information Systems Research, 2024, Vol 35, Issue 3, p1382

1526-5536

Academic Journal

10.1287/isre.2020.0277