{"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/constructing-narrative-event-evolutionary","title":"Constructing Narrative Event Evolutionary Graph for Script Event Prediction","arxiv_id":"1805.05081","date":"2018-05-14","proceeding":null,"authors":["Zhongyang Li","Xiao Ding","Ting Liu"],"abstract":"Script event prediction requires a model to predict the subsequent event\ngiven an existing event context. Previous models based on event pairs or event\nchains cannot make full use of dense event connections, which may limit their\ncapability of event prediction. To remedy this, we propose constructing an\nevent graph to better utilize the event network information for script event\nprediction. In particular, we first extract narrative event chains from large\nquantities of news corpus, and then construct a narrative event evolutionary\ngraph (NEEG) based on the extracted chains. NEEG can be seen as a knowledge\nbase that describes event evolutionary principles and patterns. To solve the\ninference problem on NEEG, we present a scaled graph neural network (SGNN) to\nmodel event interactions and learn better event representations. Instead of\ncomputing the representations on the whole graph, SGNN processes only the\nconcerned nodes each time, which makes our model feasible to large-scale\ngraphs. By comparing the similarity between input context event representations\nand candidate event representations, we can choose the most reasonable\nsubsequent event. Experimental results on widely used New York Times corpus\ndemonstrate that our model significantly outperforms state-of-the-art baseline\nmethods, by using standard multiple choice narrative cloze evaluation.","url_abs":"http://arxiv.org/abs/1805.05081v2","url_pdf":"http://arxiv.org/pdf/1805.05081v2.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":"constructing-narrative-event-evolutionary","repo_url":"https://github.com/eecrazy/ConstructingNEEG_IJCAI_2018","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"graph-neural-network","task_name":"Graph Neural Network"},{"task_slug":"multiple-choice","task_name":"Multiple-choice"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.05081","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}