• 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 : 29 / 32

ÇѱÛÁ¦¸ñ(Korean Title) ÅؽºÆ® ä¿ì±â¿Í Àû´ë ½Å°æ¸ÁÀ» ÀÌ¿ëÇÑ °³Ã¼¸í ÀÎ½Ä µ¥ÀÌÅÍ È®Àå
¿µ¹®Á¦¸ñ(English Title) Automatic Data Augmentation for Named Entity Recognition using a Text Infilling technique and Generative Adversarial Network
ÀúÀÚ(Author) ¹Úõ¿ë   ÀÌ°øÁÖ   Cheon-Young Park   Kong Joo Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 48 NO. 04 PP. 0462 ~ 0468 (2021. 04)
Çѱ۳»¿ë
(Korean Abstract)
ÀÚ¿¬ ¾ð¾î 󸮿¡ µö·¯´× ¸ðµ¨ÀÌ Àû¿ëµÇ¸é¼­ µö·¯´× ¸ðµ¨À» ±¸ÃàÇϱâ À§ÇØ ¸¹Àº ¾çÀÇ µ¥ÀÌÅÍ °¡ ÇÊ¿äÇØÁ³´Ù. ±×·¯³ª °³Ã¼¸í Àνİú °°ÀÌ ·¹ÀÌºí¸µµÈ ÇнÀ µ¥ÀÌÅÍ ±¸ÃàÀº ¾î·Á¿ö µ¥ÀÌÅÍ ºÎÁ· ¹®Á¦°¡ ¹ß»ýÇÑ´Ù. ÀÌ·¯ÇÑ µ¥ÀÌÅÍ ºÎÁ· ¹®Á¦¸¦ ÇØ°áÇϱâ À§ÇØ µ¥ÀÌÅÍ È®ÀåÀÌ ÇÊ¿äÇÏ´Ù. µû¶ó¼­ º» ¿¬±¸¿¡¼­´Â ÅؽºÆ® ä¿ì±â¿Í »ý¼ºÀû Àû´ë ½Å°æ¸ÁÀ» ÀÌ¿ëÇØ ·¹ÀÌºí¸µµÈ °³Ã¼¸í ÀÎ½Ä µ¥ÀÌÅÍ È®Àå¸ðµ¨À» Á¦¾ÈÇÑ´Ù. Á¦¾ÈÇÑ ¸ðµ¨Àº °³Ã¼¸í Á¤º¸¸¦ º¯°æÇÏÁö ¾Ê°í ºÎºÐ ¹®ÀåÀ» »ý¼ºÇØ »õ·Î¿î µ¥ÀÌÅ͸¦ »ý¼ºÇÒ ¼ö ÀÖ´Ù. Á¦¾ÈÇÑ ¸ðµ¨Àº ´Ù¸¥ ºñ±³ ¸ðµ¨µé¿¡ ºñÇØ ÀÚ¿¬½º·´°í ³»¿ëÀû ÀÏ°ü¼ºÀÌ ÀÖ´Â ºÎºÐ ¹®ÀåÀ» »ý¼ºÇÒ ¼ö ÀÖ´Ù. ¶ÇÇÑ Á¦¾ÈÇÑ ¸ðµ¨·Î È®ÀåÇÑ °³Ã¼¸í ÀÎ½Ä µ¥ÀÌÅÍ·Î °³Ã¼¸í ÀÎ½Ä ¸ðµ¨À» ÇнÀÇÒ °æ¿ì ¼º´ÉÀ» Çâ»óµÉ ¼ö ÀÖÀ½À» º¸¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
Deep neural networks have been widely used in many NLP applications, However, successful construction of deep networks requires a large training corpus. Collecting a large training corpus that contains label information such as named entities is difficult and leads to a lack of data. Automatic data augmentation represents a solution to data scarcity problem. In this paper, we propose an automatic data augmentation technique for named entity recognition(NER) based on a text infilling model and generative adversarial networks. A text infilling model is used to fill missing components of a template to generate complete sentences. Using the text infilling model, we can fill in the blank of the template to generate complete and semantically coherence text with accurately named entity labels. Sentences generated by our model show lower perplexity and higher diversity than those generated in the previous approaches. Also text augmentation based on our model can improve the performance of a conventional NER system.
Å°¿öµå(Keyword) °³Ã¼¸í ÀνĠ  µ¥ÀÌÅÍ È®Àå   ÅؽºÆ® ä¿ì±â   »ý¼ºÀû Àû´ë ½Å°æ¸Á   named entity recognition   data augmentation   text infilling   generative adversarial network  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå