{"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/integrating-local-context-and-global","title":"Integrating Local Context and Global Cohesiveness for Open Information Extraction","arxiv_id":"1804.09931","date":"2018-04-26","proceeding":null,"authors":["Qi Zhu","Xiang Ren","Jingbo Shang","Yu Zhang","Ahmed El-Kishky","Jiawei Han"],"abstract":"Extracting entities and their relations from text is an important task for\nunderstanding massive text corpora. Open information extraction (IE) systems\nmine relation tuples (i.e., entity arguments and a predicate string to describe\ntheir relation) from sentences. These relation tuples are not confined to a\npredefined schema for the relations of interests. However, current Open IE\nsystems focus on modeling local context information in a sentence to extract\nrelation tuples, while ignoring the fact that global statistics in a large\ncorpus can be collectively leveraged to identify high-quality sentence-level\nextractions. In this paper, we propose a novel Open IE system, called ReMine,\nwhich integrates local context signals and global structural signals in a\nunified, distant-supervision framework. Leveraging facts from external\nknowledge bases as supervision, the new system can be applied to many different\ndomains to facilitate sentence-level tuple extractions using corpus-level\nstatistics. Our system operates by solving a joint optimization problem to\nunify (1) segmenting entity/relation phrases in individual sentences based on\nlocal context; and (2) measuring the quality of tuples extracted from\nindividual sentences with a translating-based objective. Learning the two\nsubtasks jointly helps correct errors produced in each subtask so that they can\nmutually enhance each other. Experiments on two real-world corpora from\ndifferent domains demonstrate the effectiveness, generality, and robustness of\nReMine when compared to state-of-the-art open IE systems.","url_abs":"http://arxiv.org/abs/1804.09931v4","url_pdf":"http://arxiv.org/pdf/1804.09931v4.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":"integrating-local-context-and-global","repo_url":"https://github.com/GentleZhu/ReMine","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"open-information-extraction","task_name":"Open Information Extraction"},{"task_slug":null,"task_name":"Relation"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}