{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/personalized-video-recommendation-using-rich","title":"Personalized Video Recommendation Using Rich Contents from Videos","arxiv_id":"1612.06935","date":"2016-12-21","proceeding":null,"authors":["Xingzhong Du","Hongzhi Yin","Ling Chen","Yang Wang","Yi Yang","Xiaofang Zhou"],"abstract":"Video recommendation has become an essential way of helping people explore\nthe massive videos and discover the ones that may be of interest to them. In\nthe existing video recommender systems, the models make the recommendations\nbased on the user-video interactions and single specific content features. When\nthe specific content features are unavailable, the performance of the existing\nmodels will seriously deteriorate. Inspired by the fact that rich contents\n(e.g., text, audio, motion, and so on) exist in videos, in this paper, we\nexplore how to use these rich contents to overcome the limitations caused by\nthe unavailability of the specific ones. Specifically, we propose a novel\ngeneral framework that incorporates arbitrary single content feature with\nuser-video interactions, named as collaborative embedding regression (CER)\nmodel, to make effective video recommendation in both in-matrix and\nout-of-matrix scenarios. Our extensive experiments on two real-world\nlarge-scale datasets show that CER beats the existing recommender models with\nany single content feature and is more time efficient. In addition, we propose\na priority-based late fusion (PRI) method to gain the benefit brought by the\nintegrating the multiple content features. The corresponding experiment shows\nthat PRI brings real performance improvement to the baseline and outperforms\nthe existing fusion methods.","url_abs":"http://arxiv.org/abs/1612.06935v6","url_pdf":"http://arxiv.org/pdf/1612.06935v6.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"personalized-video-recommendation-using-rich","repo_url":"https://github.com/domainxz/top-k-rec","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"recommendation-systems","task_name":"Recommendation Systems"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}