• 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 :

ÇѱÛÁ¦¸ñ(Korean Title) ÅؽºÆ® ¹Ù²ã ¾²±â °úÁ¦¸¦ À§ÇÑ ºÐ·ù ¸ðµ¨ ±â¹ÝÀÇ ¼Õ½Ç ÇÔ¼ö ¼³°è¿Í Æò°¡
¿µ¹®Á¦¸ñ(English Title) Design and Evaluation of Loss Functions based on Classification Models
ÀúÀÚ(Author) ±è°æÈÆ   ¹ÚÁø¿í   ÀÌÁöÀº   ¹Ú»óÇö   Kyeonghun Kim   Jinuk Park   Jieun Lee   Sanghyun Park   ÀüÇö±Ô   Á¤À±°æ   Hyun-Kyu Jeon   Yun-Gyung Cheong  
¿ø¹®¼ö·Ïó(Citation) VOL 48 NO. 10 PP. 1132 ~ 1141 (2021. 10)
Çѱ۳»¿ë
(Korean Abstract)
¹Ù²ã ¾²±â(paraphrase generation)´Â ÀÔ·Â ¹®Àå¿¡ ´ëÇÏ¿© Àǹ̴ °°Áö¸¸, ´Ü¾î³ª Åë»ç ±¸Á¶¿Í °°Àº Ç¥ÇöÀÌ ´Ù¸¥ ¹®ÀåÀ» »ý¼ºÇÏ´Â °úÁ¦ÀÌ´Ù. ÃÖ±Ù À̸¦ ±¸ÇöÇϱâ À§ÇØ Àΰø ½Å°æ¸Á ±â¹ÝÀÇ ¸ðµ¨ÀÌ ³Î¸® »ç¿ëµÇ¸ç, ÇнÀ ¹æ¹ýÀ¸·Î¼­ Áöµµ ÇнÀÀÌ ÁÖ·Î »ç¿ëµÈ´Ù. ±×·¯³ª »ý¼ºµÈ ¹®Àå°ú ·¹ÀÌºí ¹®Àå °£ÀÇ Â÷À̸¦ ÁÙÀÌ´Â Áöµµ ÇнÀ ¹æ¹ýÀº ¸ðµ¨¿¡ Á¦ÇÑµÈ ÀÇ¹Ì Á¤º¸¸¸À» Á¦°øÇÑ´Ù. µû¶ó¼­ º» ³í¹®¿¡¼­´Â ºÐ·ù °úÁ¦¸¦ ÇнÀÇÑ º°µµÀÇ ¸ðµ¨À» È°¿ëÇÏ¿©, ¹Ù²ã ¾²±â ¸ðµ¨ ÇнÀ ½Ã ÀÇ¹Ì Á¤º¸¸¦ ÃßÃâÇÏ°í À̸¦ È°¿ëÇÏ´Â ¹æ¹ýÀ» Á¦¾ÈÇÏ°í ½ÇÇèÇÏ¿´À¸¸ç, ±× °á°ú ±âÁ¸ ¹æ¹ý°ú ºñ±³ÇÏ¿© ´õ ÁÁÀº ¼º´ÉÀ» º¸¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
Paraphrase generation is a task in which the model generates an output sentence conveying the same meaning as the given input text but with a different representation. Recently, paraphrase generation has been widely used for solving the task of using artificial neural networks with supervised learning between the model¡¯s prediction and labels. However, this method gives limited information because it only detects the representational difference. For that reason, we propose a method to extract semantic information with classification models and use them for the training loss function. Our evaluations showed that the proposed method outperformed baseline models.
Å°¿öµå(Keyword) ÀÚ¿¬¾î 󸮠  ¸ÖƼ ¸ð´Þ   °¨¼º ºÐ¼®   °¨Á¤ ŽÁö   ´ÜÀÏ ÀÔÃâ·Â Æ®·£½ºÆ÷¸Ó   BERT   natural language processing   multimodal   sentiment analysis   emotion detection   singlestream transformer   BERT   ¹Ù²ã ¾²±â   ºÐ·ù ¸ðµ¨   ¸ðµ¨±â¹Ý ÇнÀ   ¼Õ½Ç ÇÔ¼ö   paraphrase generation   classification model   model-based learning   loss function  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå