{"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/chinese-lexical-analysis-with-deep-bi-gru-crf","title":"Chinese Lexical Analysis with Deep Bi-GRU-CRF Network","arxiv_id":"1807.01882","date":"2018-07-05","proceeding":null,"authors":["Zhenyu Jiao","Shuqi Sun","Ke Sun"],"abstract":"Lexical analysis is believed to be a crucial step towards natural language\nunderstanding and has been widely studied. Recent years, end-to-end lexical\nanalysis models with recurrent neural networks have gained increasing\nattention. In this report, we introduce a deep Bi-GRU-CRF network that jointly\nmodels word segmentation, part-of-speech tagging and named entity recognition\ntasks. We trained the model using several massive corpus pre-tagged by our best\nChinese lexical analysis tool, together with a small, yet high-quality human\nannotated corpus. We conducted balanced sampling between different corpora to\nguarantee the influence of human annotations, and fine-tune the CRF decoding\nlayer regularly during the training progress. As evaluated by linguistic\nexperts, the model achieved a 95.5% accuracy on the test set, roughly 13%\nrelative error reduction over our (previously) best Chinese lexical analysis\ntool. The model is computationally efficient, achieving the speed of 2.3K\ncharacters per second with one thread.","url_abs":"http://arxiv.org/abs/1807.01882v1","url_pdf":"http://arxiv.org/pdf/1807.01882v1.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":"chinese-lexical-analysis-with-deep-bi-gru-crf","repo_url":"https://github.com/baidu/lac","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"paddle","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"chinese-lexical-analysis-with-deep-bi-gru-crf","repo_url":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/examples/lexical_analysis","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":null}],"tasks":[{"task_slug":"lexical-analysis","task_name":"Lexical Analysis"},{"task_slug":"named-entity-recognition-1","task_name":"Named Entity Recognition"},{"task_slug":"named-entity-recognition-ner","task_name":"Named Entity Recognition (NER)"},{"task_slug":"natural-language-understanding","task_name":"Natural Language Understanding"},{"task_slug":"part-of-speech-tagging","task_name":"Part-Of-Speech Tagging"},{"task_slug":"named-entity-recognition","task_name":"named-entity-recognition"}],"methods":[{"method_slug":"crf","method_name":"CRF"},{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1807.01882","atlas_url":"https://app.syntology.ai/?focus=1807.01882","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}