Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
SMERT: °¨¼º ºÐ¼® ¹× °¨Á¤ ŽÁö¸¦ À§ÇÑ ´ÜÀÏ ÀÔÃâ·Â ¸ÖƼ ¸ð´Þ BERT |
¿µ¹®Á¦¸ñ(English Title) |
SMERT: Single-stream Multimodal BERT for Sentiment Analysis and Emotion Detection |
ÀúÀÚ(Author) |
¹®È¿Á¾
¼Õ½Ã¿î
¹®¾ç¼¼
Hyojong Moon
Siwoon Son
Yang-Sae Moon
±è°æÈÆ
¹ÚÁø¿í
ÀÌÁöÀº
¹Ú»óÇö
Kyeonghun Kim
Jinuk Park
Jieun Lee
Sanghyun Park
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 48 NO. 10 PP. 1122 ~ 1131 (2021. 10) |
Çѱ۳»¿ë (Korean Abstract) |
°¨¼º ºÐ¼®Àº ÅؽºÆ®·ÎºÎÅÍ ÁÖ°üÀûÀÎ ÀÇ°ß ¹× ¼ºÇâÀ» ºÐ¼®ÇÏ°í, °¨Á¤ ŽÁö´Â 'Çູ', '½½ÇÄ'°ú °°ÀÌ ÅؽºÆ®¿¡¼ ³ªÅ¸³ª´Â °¨Á¤À» °ËÃâÇÏ´Â ¿¬±¸´Ù. ¸ÖƼ ¸ð´Þ µ¥ÀÌÅÍ´Â ÅؽºÆ®»Ó¸¸ ¾Æ´Ï¶ó À̹ÌÁö, À½¼º µ¥ÀÌÅÍ°¡ ÇÔ²² ³ªÅ¸³ª´Â °ÍÀ» ÀǹÌÇÑ´Ù. °ü·Ã ¼±Çà ¿¬±¸¿¡¼ ¼øȯ ½Å°æ¸Á ¸ðÇü ȤÀº ±³Â÷ Æ®·£½ºÆ÷¸Ó¸¦ »ç¿ëÇÑ´Ù. ÇÏÁö¸¸ ¼øȯ ½Å°æ¸Á ¸ðÇüÀº Àå±â ÀÇÁ¸¼º ¹®Á¦¸¦ °¡Áö¸ç, ±³Â÷ Æ®·£½ºÆ÷¸Ó´Â ¸ð´Þ¸®Æ¼º° Ư¼ºÀ» ¹Ý¿µÇÏÁö ¸øÇÏ´Â ¹®Á¦Á¡ÀÌ ÀÖ´Ù. À̸¦ ÇØ°áÇϱâ À§ÇØ º» ¿¬±¸¿¡¼´Â ¸ÖƼ ¸ð´Þ µ¥ÀÌÅÍ°¡ ÇϳªÀÇ ³×Æ®¿öÅ©·Î ÇнÀµÇ´Â ´ÜÀÏ ÀÔÃâ·Â Æ®·£½ºÆ÷¸Ó ±â¹Ý ¸ðÇü SMERT¸¦ Á¦¾ÈÇÑ´Ù. SMERT´Â ¸ð´Þ¸®Æ¼ °áÇÕ Ç¥ÇöÇüÀ» ¾ò¾î À̸¦ °¨¼º ºÐ¼® ¹× °¨Á¤ ŽÁö¿¡ È°¿ëÇÑ´Ù. ¶ÇÇÑ, BERTÀÇ ÈƷà ŽºÅ©¸¦ ¸ÖƼ ¸ð´Þ µ¥ÀÌÅÍ¿¡ È°¿ëÇϱâ À§ÇØ °³·®ÇÏ¿© »ç¿ëÇÑ´Ù. Á¦¾ÈÇÏ´Â ¸ðµ¨ÀÇ °ËÁõÀ» À§ÇØ CMU-MOSEI µ¥ÀÌÅͼ°ú ¿©·¯ Æò°¡ ÁöÇ¥¸¦ ÀÌ¿ëÇÏ°í, ¸ð´Þ¸®Æ¼ Á¶ÇÕº° ºñ±³½ÇÇè°ú ¿¹½Ã¸¦ ÅëÇØ ¸ðµ¨ÀÇ ¿ì¼ö¼ºÀ» °ËÁõÇÏ¿´´Ù. |
¿µ¹®³»¿ë (English Abstract) |
Sentiment Analysis is defined as a task that analyzes subjective opinion or propensity and, Emotion Detection is the task that finds emotions such as 'happy' or 'sad' from text data. Multimodal data refers to the appearance of image and voice data in addition to text data. In prior research, RNN or cross-transformer models were used, however, RNN models have long-term dependency problems. Also, since cross-transformer models could not capture the attribute of modalities, they got worse results. To solve those problems, we propose SMERT based on a single-stream transformer ran on a single network. SMERT can get joint representation for Sentiment Analysis and Emotion Detection. Besides, we use BERT tasks which are improved to utilize for multimodal data. To present the proposed model, we verify the superiority of SMERT through a comparative experiment on the combination of modalities using the CMU-MOSEI dataset and various evaluation metrics. |
Å°¿öµå(Keyword) |
µ¥ÀÌÅÍ ½ºÆ®¸²
µö·¯´× Ãß·Ð
½ºÅÂÅ·
ºÐ»ê ó¸®
¾ÆÆÄÄ¡ ½ºÅè
data stream
deep learning inference
stacking
distributed processing
Apache Storm
ÀÚ¿¬¾î ó¸®
¸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
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|