Latent User Intent Modeling for Sequential Recommenders
This addresses the need for better user intent modeling in industrial recommender systems to optimize long-term user experience, though it appears incremental by applying existing VAE methods to this domain.
The paper tackled the problem of sequential recommenders lacking higher-level understanding of user intents by proposing a probabilistic modeling approach with latent variables inferred using variational autoencoders, and demonstrated effectiveness through offline analyses and live experiments on a large-scale industrial platform.
Sequential recommender models are essential components of modern industrial recommender systems. These models learn to predict the next items a user is likely to interact with based on his/her interaction history on the platform. Most sequential recommenders however lack a higher-level understanding of user intents, which often drive user behaviors online. Intent modeling is thus critical for understanding users and optimizing long-term user experience. We propose a probabilistic modeling approach and formulate user intent as latent variables, which are inferred based on user behavior signals using variational autoencoders (VAE). The recommendation policy is then adjusted accordingly given the inferred user intent. We demonstrate the effectiveness of the latent user intent modeling via offline analyses as well as live experiments on a large-scale industrial recommendation platform.