Recommender System
Recommender systems aim to predict user preferences and provide personalized recommendations, enhancing user experience across various online platforms. Current research emphasizes improving accuracy and mitigating biases, focusing on advanced techniques like neural networks (including transformers and recurrent networks), matrix factorization, and ensemble methods to address challenges such as data sparsity, outlier detection, and the impact of algorithmic bias on user preferences. This field is significant due to its widespread applications and the growing need for responsible and ethical design, driving research into explainability, fairness, and the use of causal inference to understand and mitigate the societal impact of these systems.
Papers
Language-Model Prior Overcomes Cold-Start Items
Shiyu Wang, Hao Ding, Yupeng Gu, Sergul Aydore, Kousha Kalantari, Branislav Kveton
Recommender systems and reinforcement learning for human-building interaction and context-aware support: A text mining-driven review of scientific literature
Wenhao Zhang, Matias Quintana, Clayton Miller