Abstract
AbstractPersonalized news headline generation aims to model users interests through their news browsing history, and then generate news headlines based on their preferences. Using the user representation of existing news recommendation systems to personalize news headlines has been very successful, but there are still many problems to be solved. Firstly, news recommendation and headline generation tasks have different requirements for user interest modelling. Secondly, the user interest embedding approach does not consider the redundancy of user interests and the complementarity between user interests and candidate news texts. In this paper, we propose a new personalized headline generation framework to enhance the user interest perception. It can highlight user interests related to candidate texts for accurate embedding. We further offer to improve user interest representation using entity words in the news to meet the demand for news headline personalization. Extensive experiments show that our approach achieves better performance in headline generation tasks.KeywordsPersonalized headline generationNews recommendationReinforcement learning
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
