{"id":553,"date":"2020-09-30T11:08:00","date_gmt":"2020-09-30T03:08:00","guid":{"rendered":"http:\/\/140.115.70.194:6417\/?p=553"},"modified":"2024-11-09T19:15:50","modified_gmt":"2024-11-09T11:15:50","slug":"gated-graph-sequence-neural-networks-for-chinese-healthcare-named-entity-recognition","status":"publish","type":"post","link":"https:\/\/ainlp.tw\/index.php\/2020\/09\/30\/gated-graph-sequence-neural-networks-for-chinese-healthcare-named-entity-recognition\/","title":{"rendered":"\u9580\u63a7\u5716\u5e8f\u5217\u795e\u7d93\u7db2\u8def\u4e4b\u4e2d\u6587\u5065\u5eb7\u7167\u8b77\u547d\u540d\u5be6\u9ad4\u8fa8\u8b58"},"content":{"rendered":"\n<p>\u76e7\u6bc5\u3001\u674e\u9f8d\u8c6a\u3002<\/p>\n\n\n\n<p>In\u00a0<em>Proceedings of the 32th Conference on Computational Linguistics and Speech Processing\u00a0<strong>(ROCLING\u201920)<\/strong>, <\/em>pages 21-36<em>.<\/em><\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity is-style-wide\"\/>\n\n\n\n<p class=\"has-medium-font-size\"><strong>\u6458\u8981<\/strong><\/p>\n\n\n\n<p style=\"font-size:clamp(14px, 0.875rem + ((1vw - 3.2px) * 0.156), 16px);\">\u547d\u540d\u5be6\u9ad4\u8fa8\u8b58\u4efb\u52d9\u7684\u76ee\u6a19\u662f\u5f9e\u975e\u7d50\u69cb\u5316\u7684\u8f38\u5165\u6587\u672c\u4e2d\uff0c\u62bd\u53d6\u51fa\u95dc\u6ce8\u7684\u547d\u540d\u5be6\u9ad4\uff0c\u4f8b\u5982: \u4eba\u540d\u3001\u5730\u540d\u3001\u7d44\u7e54\u540d\u3001\u65e5\u671f\u3001\u6642\u9593\u7b49\u5c08\u6709\u540d\u8a5e\uff0c\u64f7\u53d6\u7684\u547d\u540d\u5be6\u9ad4\uff0c\u53ef\u4ee5\u505a\u70ba\u95dc\u4fc2\u64f7\u53d6\u3001 \u4e8b\u4ef6\u5075\u6e2c\u8207\u8ffd\u8e64\u3001\u77e5\u8b58\u5716\u8b5c\u5efa\u7f6e\u3001\u554f\u7b54\u7cfb\u7d71\u7b49\u61c9\u7528\u7684\u57fa\u790e\u3002\u6a5f\u5668\u5b78\u7fd2\u7684\u65b9\u6cd5\u5c07\u5176\u8996\u70ba\u5e8f \u5217\u6a19\u8a3b\u554f\u984c\uff0c\u900f\u904e\u5927\u898f\u6a21\u8a9e\u6599\u5b78\u7fd2\u6a19\u8a3b\u6a21\u578b\uff0c\u5c0d\u53e5\u5b50\u7684\u5404\u500b\u5b57\u5143\u4f4d\u7f6e\u9032\u884c\u6a19\u8a3b\u3002\u6211\u5011\u63d0 \u51fa\u4e00\u500b\u9580\u63a7\u5716\u5e8f\u5217\u795e\u7d93\u7db2\u8def&nbsp;(Gated Graph Sequence Neural Network, GGSNN)&nbsp;\u6a21\u578b\uff0c\u7528 \u65bc\u4e2d\u6587\u5065\u5eb7\u7167\u8b77\u9818\u57df\u547d\u540d\u5be6\u9ad4\u8fa8\u8b58\uff0c\u6211\u5011\u6574\u5408\u8a5e\u5d4c\u5165\u4ee5\u53ca\u90e8\u9996\u5d4c\u5165\u7684\u8cc7\u8a0a\uff0c\u5efa\u69cb\u591a\u91cd\u5d4c \u5165\u7684\u5b57\u5d4c\u5165\u5411\u91cf\uff0c\u85c9\u7531\u8abf\u9069\u9580\u63a7\u5716\u5e8f\u5217\u795e\u7d93\u7db2\u8def\uff0c\u878d\u5165\u5df2\u77e5\u5b57\u5178\u4e2d\u7684\u547d\u540d\u5be6\u9ad4\u8cc7\u8a0a\uff0c\u7136 \u5f8c\u929c\u63a5\u96d9\u5411\u9577\u77ed\u671f\u8a18\u61b6\u985e\u795e\u7d93\u7db2\u8def\u8207\u689d\u4ef6\u96a8\u6a5f\u5834\u57df\uff0c\u5c0d\u4e2d\u6587\u53e5\u5b50\u4e2d\u7684\u5b57\u5143\u5e8f\u5217\u6a19\u8a3b\u3002\u6211 \u5011\u900f\u904e\u7db2\u8def\u722c\u87f2\u8490\u96c6\u5065\u5eb7\u7167\u8b77\u76f8\u95dc\u5167\u5bb9\u7684\u7db2\u8def\u6587\u7ae0\u4ee5\u53ca\u91ab\u7642\u554f\u7b54\u7d00\u9304\uff0c\u7136\u5f8c\u96a8\u6a5f\u62bd\u53d6\u4e2d \u6587\u53e5\u5b50\u505a\u4eba\u5de5\u65b7\u8a5e\u8207\u547d\u540d\u5be6\u9ad4\u6a19\u8a18\uff0c\u53e5\u5b50\u7e3d\u6578\u70ba&nbsp;30,692&nbsp;\u53e5&nbsp;(\u7d04&nbsp;150&nbsp;\u842c\u5b57\/91.7&nbsp;\u842c\u8a5e)\uff0c\u5171 \u6709&nbsp;68,460&nbsp;\u547d\u540d\u5be6\u9ad4\uff0c\u5305\u542b&nbsp;10&nbsp;\u500b\u547d\u540d\u5be6\u9ad4\u7a2e\u985e:\u4eba\u9ad4\u3001\u75c7\u72c0\u3001\u91ab\u7642\u5668\u6750\u3001\u6aa2\u9a57\u3001\u5316\u5b78\u7269 \u8cea\u3001\u75be\u75c5\u3001\u85e5\u54c1\u3001\u71df\u990a\u54c1\u3001\u6cbb\u7642\u8207\u6642\u9593\u3002\u85c9\u7531\u5be6\u9a57\u7d50\u679c\u8207\u932f\u8aa4\u5206\u6790\u5f97\u77e5\uff0c\u6211\u5011\u63d0\u51fa\u7684\u6a21 \u578b\u9054\u5230\u6700\u597d\u7684&nbsp;F1-score 75.69%\uff0c\u6bd4\u76f8\u95dc\u7814\u7a76\u6a21\u578b&nbsp;(BiLSTM-CRF, BERT, Lattice, Gazetteers&nbsp;\u4ee5\u53ca&nbsp;ME-CNER)\u8868\u73fe\u597d\uff0c\u4e14\u70ba\u6548\u80fd\u8207\u6548\u7387\u517c\u5177\u7684\u4e2d\u6587\u5065\u5eb7\u7167\u8b77\u547d\u540d\u5be6\u9ad4\u8fa8\u8b58 \u65b9\u6cd5\u3002<\/p>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u76e7\u6bc5\u3001\u674e\u9f8d\u8c6a<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_themeisle_gutenberg_block_has_review":false,"footnotes":""},"categories":[6],"tags":[40,44,13],"class_list":["post-553","post","type-post","status-publish","format-standard","hentry","category-achievements","tag-40","tag-healthcare","tag-rocling"],"blocksy_meta":[],"_links":{"self":[{"href":"https:\/\/ainlp.tw\/index.php\/wp-json\/wp\/v2\/posts\/553","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/ainlp.tw\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/ainlp.tw\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/ainlp.tw\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/ainlp.tw\/index.php\/wp-json\/wp\/v2\/comments?post=553"}],"version-history":[{"count":2,"href":"https:\/\/ainlp.tw\/index.php\/wp-json\/wp\/v2\/posts\/553\/revisions"}],"predecessor-version":[{"id":815,"href":"https:\/\/ainlp.tw\/index.php\/wp-json\/wp\/v2\/posts\/553\/revisions\/815"}],"wp:attachment":[{"href":"https:\/\/ainlp.tw\/index.php\/wp-json\/wp\/v2\/media?parent=553"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/ainlp.tw\/index.php\/wp-json\/wp\/v2\/categories?post=553"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/ainlp.tw\/index.php\/wp-json\/wp\/v2\/tags?post=553"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}