Personalized promotion recommendation: A dynamic adaptation modeling approach

The targeting and timing of sales promotions have long been critical challenges in marketing efforts. While traditional strategies concentrate on customer segmentation analysis, the enriched data environment in e-commerce nowadays calls for sophisticated analytical approaches that enable sellers to target their promotional activities at the personalized level. Meanwhile, although research on recommender systems has accumulated abundant techniques for profiling personalized preferences, the dynamic influences of sales promotions on consumer preferences remain underexplored. Therefore, we propose that personalized product recommendations can be consolidated with consumers' evolving perceptions of sales promotions. Drawing upon transaction utility theory and adaptation level theory, this paper develops a novel recommendation approach DAMA, in which consumer adaptation factor, is introduced to model the temporal dynamics of consumers' preferences and price perceptions simultaneously in the context of diverse sales promotional activities. Extensive empirical experiments are conducted on real-world datasets against baseline methods, revealing the performance superiority of DAMA.