Context-based Fast Recommendation Strategy for Long User Behavior Sequence in Meituan Waimai

Zhichao Feng, Junjie Xie, Kaiyuan Li, Yu Qin, Pengfei Wang, Qianzhong Li, Bin Yin, Xiang Li, Wei Lin, Shangguang Wang

Abstract: In the recommender system of Meituan Waimai, we are dealing with ever-lengthening user behavior sequences, which pose an increasing challenge to modeling user preference effectively. A number of existing sequential recommendation models struggle to capture long-term dependencies, or they exhibit high complexity, both of which make it difficult to satisfy the unique business requirements of Meituan Waimai's recommender system.To better model user interests, we consider selecting relevant sub-sequences from users' extensive historical behaviors based on their preferences. In this specific scenario, we've noticed that the contexts in which users interact have a significant impact on their preferences. For this purpose, we introduce a novel method called Context-based Fast Recommendation Strategy (referred to as CoFARS) to tackle the issue of long sequences. We first identify contexts that share similar user preferences with the target context and then locate the corresponding Points of Interest (PoIs) based on these identified contexts. This approach eliminates the necessity to select a sub-sequence for every candidate PoI, thereby avoiding high time complexity. Specifically, we implement a prototype-based approach to pinpoint contexts that mirror similar user preferences. To amplify accuracy and interpretability, we employ Jensen?Shannon(JS) divergence of PoI attributes such as categories and prices as a measure of similarity between contexts. Subsequently, we construct a temporal graph that encompasses both prototype and context nodes to integrate temporal information. We then identify appropriate prototypes considering both target contexts and short-term user preferences. Following this, we utilize contexts aligned with these prototypes to generate a sub-sequence, aimed at predicting CTR and CTCVR scores with target attention.Since its inception in 2023, this strategy has been adopted in Meituan Waimai's display recommender system, leading to a 4.6% surge in CTR and a 4.2% boost in GMV.