{"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/dual-graph-attention-networks-for-deep-latent","title":"Dual Graph Attention Networks for Deep Latent Representation of Multifaceted Social Effects in Recommender Systems","arxiv_id":"1903.10433","date":"2019-03-25","proceeding":null,"authors":["Qitian Wu","Hengrui Zhang","Xiaofeng Gao","Peng He","Paul Weng","Han Gao","Guihai Chen"],"abstract":"Social recommendation leverages social information to solve data sparsity and\ncold-start problems in traditional collaborative filtering methods. However,\nmost existing models assume that social effects from friend users are static\nand under the forms of constant weights or fixed constraints. To relax this\nstrong assumption, in this paper, we propose dual graph attention networks to\ncollaboratively learn representations for two-fold social effects, where one is\nmodeled by a user-specific attention weight and the other is modeled by a\ndynamic and context-aware attention weight. We also extend the social effects\nin user domain to item domain, so that information from related items can be\nleveraged to further alleviate the data sparsity problem. Furthermore,\nconsidering that different social effects in two domains could interact with\neach other and jointly influence user preferences for items, we propose a new\npolicy-based fusion strategy based on contextual multi-armed bandit to weigh\ninteractions of various social effects. Experiments on one benchmark dataset\nand a commercial dataset verify the efficacy of the key components in our\nmodel. The results show that our model achieves great improvement for\nrecommendation accuracy compared with other state-of-the-art social\nrecommendation methods.","url_abs":"http://arxiv.org/abs/1903.10433v1","url_pdf":"http://arxiv.org/pdf/1903.10433v1.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":"dual-graph-attention-networks-for-deep-latent","repo_url":"https://github.com/echo740/DANSER-WWW-19","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"collaborative-filtering","task_name":"Collaborative Filtering"},{"task_slug":"graph-attention","task_name":"Graph Attention"},{"task_slug":"recommendation-systems","task_name":"Recommendation Systems"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/recommendation-systems-on-epinions","task":"Recommendation Systems","dataset":"Epinions","model":"DANSER","rank_in_archive_order":1,"of":4,"metrics":{"MAE":"0.7781","RMSE":"1.0268"},"uses_additional_data":false},{"leaderboard":"/sota/recommendation-systems-on-wechat","task":"Recommendation Systems","dataset":"WeChat","model":"DANSER","rank_in_archive_order":1,"of":2,"metrics":{"AUC":"0.8165","P@10":"0.0823"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1903.10433","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}