{"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/neural-relation-extraction-via-inner-sentence","title":"Neural Relation Extraction via Inner-Sentence Noise Reduction and Transfer Learning","arxiv_id":"1808.06738","date":"2018-08-21","proceeding":"EMNLP 2018 10","authors":["Tianyi Liu","Xinsong Zhang","Wanhao Zhou","Weijia Jia"],"abstract":"Extracting relations is critical for knowledge base completion and\nconstruction in which distant supervised methods are widely used to extract\nrelational facts automatically with the existing knowledge bases. However, the\nautomatically constructed datasets comprise amounts of low-quality sentences\ncontaining noisy words, which is neglected by current distant supervised\nmethods resulting in unacceptable precisions. To mitigate this problem, we\npropose a novel word-level distant supervised approach for relation extraction.\nWe first build Sub-Tree Parse(STP) to remove noisy words that are irrelevant to\nrelations. Then we construct a neural network inputting the sub-tree while\napplying the entity-wise attention to identify the important semantic features\nof relational words in each instance. To make our model more robust against\nnoisy words, we initialize our network with a priori knowledge learned from the\nrelevant task of entity classification by transfer learning. We conduct\nextensive experiments using the corpora of New York Times(NYT) and Freebase.\nExperiments show that our approach is effective and improves the area of\nPrecision/Recall(PR) from 0.35 to 0.39 over the state-of-the-art work.","url_abs":"http://arxiv.org/abs/1808.06738v2","url_pdf":"http://arxiv.org/pdf/1808.06738v2.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":[],"tasks":[{"task_slug":"knowledge-base-completion","task_name":"Knowledge Base Completion"},{"task_slug":null,"task_name":"Relation"},{"task_slug":"relation-extraction","task_name":"Relation Extraction"},{"task_slug":"relationship-extraction-distant-supervised","task_name":"Relationship Extraction (Distant Supervised)"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/relationship-extraction-distant-supervised-on","task":"Relationship Extraction (Distant Supervised)","dataset":"New York Times Corpus","model":"BiGRU+WLA+EWA","rank_in_archive_order":8,"of":9,"metrics":{"AUC":"0.390","Average Precision":"0.390"},"uses_additional_data":false},{"leaderboard":"/sota/relationship-extraction-distant-supervised-on","task":"Relationship Extraction (Distant Supervised)","dataset":"New York Times Corpus","model":"BGRU-SET","rank_in_archive_order":9,"of":9,"metrics":{"AUC":"0.390","Average Precision":"0.390"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1808.06738","atlas_url":"https://app.syntology.ai/?focus=1808.06738","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}