{"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/cascade-dynamics-modeling-with-attention","title":"Cascade Dynamics Modeling with Attention-based Recurrent Neural Network","arxiv_id":null,"date":"2017-05-01","proceeding":null,"authors":["Yongqing Wang","HuaWei Shen","Shenghua Liu","Jinhua Gao","and Xueqi Cheng"],"abstract":"An ability of modeling and predicting the cascades\r\nof resharing is crucial to understanding informa\u0002tion propagation and to launching campaign of viral\r\nmarketing. Conventional methods for cascade pre\u0002diction heavily depend on the hypothesis of diffu\u0002sion models, e.g., independent cascade model and\r\nlinear threshold model. Recently, researchers at\u0002tempt to circumvent the problem of cascade predic\u0002tion using sequential models (e.g., recurrent neural\r\nnetwork, namely RNN) that do not require know\u0002ing the underlying diffusion model. Existing se\u0002quential models employ a chain structure to capture\r\nthe memory effect. However, for cascade predic\u0002tion, each cascade generally corresponds to a diffu\u0002sion tree, causing cross-dependence in cascade—\r\none sharing behavior could be triggered by its\r\nnon-immediate predecessor in the memory chain.\r\nIn this paper, we propose to an attention-based\r\nRNN to capture the cross-dependence in cascade.\r\nFurthermore, we introduce a coverage strategy to\r\ncombat the misallocation of attention caused by\r\nthe memoryless of traditional attention mechanism.\r\nExtensive experiments on both synthetic and real\r\nworld datasets demonstrate the proposed models\r\noutperform state-of-the-art models at both cascade\r\nprediction and inferring diffusion tree.","url_abs":"https://www.ijcai.org/Proceedings/2017/0416.pdf","url_pdf":"https://www.ijcai.org/Proceedings/2017/0416.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":"cascade-dynamics-modeling-with-attention","repo_url":"https://github.com/Allen517/cyanrnn_project","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"marketing","task_name":"Marketing"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}