• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸°úÇÐȸ ³í¹®Áö > Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Current Result Document : 7 / 17 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) ¸ÖƼÇìµå ÁÖÀÇÁýÁß ±â¹ý°ú ÇÏÀÌ¿þÀÌ ³×Æ®¿öÅ©¸¦ È°¿ëÇÑ »ý¹°ÇÐ °³Ã¼¸í ÀνÄ
¿µ¹®Á¦¸ñ(English Title) Biomedical Named Entity Recognition using Multi-head Attention with Highway Network
ÀúÀÚ(Author) Á¶¹Î¼ö   ¹ÚÁø¿í   ÇÏÁöȯ   ¹ÚÂùÈñ   ¹Ú»óÇö   Minsoo Cho   Jinuk Park   Jihwan Ha   Chanhee Park   Sanghyun Park  
¿ø¹®¼ö·Ïó(Citation) VOL 46 NO. 06 PP. 0544 ~ 0553 (2019. 06)
Çѱ۳»¿ë
(Korean Abstract)
»ý¹°ÇÐ °³Ã¼¸í ÀνÄÀ̶õ »ý¹°ÇÐ ¹®ÇåÀ¸·ÎºÎÅÍ Áúº´, À¯ÀüÀÚ, ´Ü¹éÁú°ú °°Àº »ý¹°ÇÐ °³Ã¼¸íÀ» ÃßÃâÇÏ°í ±× Á¾·ù¸¦ ºÐ·ùÇÏ´Â ÀÛ¾÷À¸·Î, »ý¹°ÇÐ µ¥ÀÌÅͷκÎÅÍ À¯ÀǹÌÇÑ Á¤º¸¸¦ ÃßÃâÇϴµ¥ Áß¿äÇÑ ¿ªÇÒÀ» ÇÑ´Ù. º» ¿¬±¸¿¡¼­´Â ÀÔ·Â ´Ü¾îÀÇ ÀÚÁúÀ» ÀÚµ¿À¸·Î ÃßÃâÇÒ ¼ö ÀÖ´Â µö·¯´× ±â¹ÝÀÇ Bi-LSTM-CRF ¸ðµ¨À» È°¿ëÇÑ °³Ã¼¸í ÀÎ½Ä ¿¬±¸¸¦ ÁøÇàÇÏ¿´´Ù. Multi-head ÁÖÀÇ ±âÁ¦ ±â¹ýÀ» Àû¿ëÇÏ¿© ÀÔ·Â ´Ü¾îµé °£ÀÇ °ü°è¸¦ Æ÷ÂøÇÏ°í °ü·Ã¼ºÀÌ ³ôÀº ´Ü¾î¿¡ ÁÖ¸ñÇÏ¿© ¿¹ÃøÀÇ ¼º´ÉÀ» ³ô¿´´Ù. ¶ÇÇÑ, ´Ü¾î ´ÜÀ§ ÀÓº£µù º¤ÅÍ ¿Ü ¹®ÀÚ ´ÜÀ§ ÀÓº£µù º¤Å͸¦ °áÇÕÇÏ¿© ÀÔ·Â ÀÓº£µùÀÇ Ç¥»óÀ» È®ÀåÇÏ°í, °¢ Ç¥»óÀÇ Á¤º¸ È帧À» ÇнÀÇϱâ À§ÇØ Highway ³×Æ®¿öÅ©¿¡ Àû¿ëÇÏ¿´´Ù. Á¦¾ÈÇÏ´Â ¸ðµ¨ÀÇ ¼º´ÉÀ» Æò°¡Çϱâ À§ÇØ µÎ °³ÀÇ ¿µ¾î »ý¹°ÇÐ µ¥ÀÌÅͼÂÀ¸·Î ºñ±³ ½ÇÇèÀ» ÁøÇàÇÏ¿´À¸¸ç, ±× °á°ú ±âÁ¸ ¿¬±¸ÀÇ ¸ðµ¨µéº¸´Ù Çâ»óµÈ ¼º´ÉÀ» º¸¿´´Ù. À̸¦ ÅëÇØ Á¦¾ÈÇÏ´Â ¹æ¹ý·ÐÀÌ »ý¹°ÇÐ °³Ã¼¸í ÀÎ½Ä ¿¬±¸¿¡¼­ È¿°úÀûÀÎ ¹æ¹ý·ÐÀÓÀ» ÀÔÁõÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
Biomedical named entity recognition(BioNER) is the process of extracting biomedical entities such as diseases, genes, proteins, and chemicals from biomedical literature. BioNER is an indispensable technique for the extraction of meaningful data from biomedical domains. The proposed model employs deep learning based Bi-LSTM-CRF model which eliminates the need for hand-crafted feature engineering. Additionally, the model contains multi-head attention to capture the relevance between words, which is used when predicting the label of each input token. Also, in the input embedding layer, the model integrates character-level embedding with word-level embedding and applies the combined word embedding into the highway network to adaptively carry each embedding to the input of the Bi-LSTM model. Two English biomedical benchmark datasets were employed in the present research to evaluate the level of performance. The proposed model resulted in higher f1-score compared to other previously studied models. The results demonstrate the effectiveness of the proposed methods in biomedical named entity recognition study.
Å°¿öµå(Keyword) Á¤º¸ ÃßÃâ   ÀÚ¿¬¾î 󸮠  °³Ã¼¸í ÀνĠ  Multi-head ÁÖÀÇ ±âÁ¦ ±â¹ý   Highway ³×Æ®¿öÅ©   ´Ü¾î ÀÓº£µù   information retrieval   natural language processing   named entity recognition   multi-head attention mechanism   highway network   word embedding  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå