Probabilistic Topic Model for Hybrid Recommender Systems: A Stochastic Variational Bayesian Approach
提出一种协变量引导的异质监督主题模型用于在线电影推荐，并开发随机变分贝叶斯框架实现大数据下的快速、可扩展和准确估计。
This paper proposes a novel covariate-guided heterogeneous supervised topic model for online movie recommendation and develops a stochastic variational Bayesian framework to achieve fast, scalable, and accurate estimation in big data settings.