This paper provides a comprehensive overview of user modeling and profiling, essential techniques in information retrieval and recommender systems for delivering personalized experiences. It traces the development from early stereotype models to the latest deep learning techniques, proposes a new taxonomy, and highlights shifts towards more sophisticated methods. The paper also emphasizes the need for privacy-preserving techniques and the push towards explainability and fairness. It explores the application of user modeling in domains like fake news detection, cybersecurity, and personalized education.
Publication date: 16 Feb 2024
Project Page: https://arxiv.org/abs/2402.09660v1
Paper: https://arxiv.org/pdf/2402.09660
