TPRM: A Topic-based Personalized Ranking Model for Web Search
This work addresses the problem of improving web search relevance for users by personalizing rankings, though it appears incremental as it builds on existing methods like pretrained representations and user profiles.
The paper tackles the challenge of designing personalized ranking systems by proposing TPRM, which integrates user topical profiles with pretrained contextualized term representations to tailor document rankings, and it demonstrates significant outperformance over state-of-the-art ad-hoc and personalized ranking models in experiments on a real-world dataset.
Ranking models have achieved promising results, but it remains challenging to design personalized ranking systems to leverage user profiles and semantic representations between queries and documents. In this paper, we propose a topic-based personalized ranking model (TPRM) that integrates user topical profile with pretrained contextualized term representations to tailor the general document ranking list. Experiments on the real-world dataset demonstrate that TPRM outperforms state-of-the-art ad-hoc ranking models and personalized ranking models significantly.