• 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 : 8 / 14 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) Parallel Stacked Bidirectional LSTM ¸ðµ¨À» ÀÌ¿ëÇÑ Çѱ¹¾î ¿µÈ­¸®ºä °¨¼º ºÐ¼®
¿µ¹®Á¦¸ñ(English Title) Korean Movie-review Sentiment Analysis Using Parallel Stacked Bidirectional LSTM Model
ÀúÀÚ(Author) ¿À¿µÅà  ±è¹ÎÅ   ±è¿ìÁÖ   Yeongtaek Oh   Mintae Kim   Wooju Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 46 NO. 01 PP. 0045 ~ 0049 (2019. 01)
Çѱ۳»¿ë
(Korean Abstract)
°¨¼ººÐ¼®Àº ÅؽºÆ® ¹®¼­ÀÇ °¨¼ºÀ» ºÐ·ùÇÏ´Â ¹®¼­ ºÐ·ùÀÇ ÇÑ ºÐ¾ßÀÌ´Ù. µö·¯´×À» ÀÌ¿ëÇÑ °¨¼º ºÐ¼® ¹æ¹ý·ÐÀº ¹®¼­¸¦ ÅäÅ«È­ ÈÄ ÀÓº£µùÀ» ÅëÇØ ¹®À庤Å͸¦ ¾ò´Â °úÁ¤°ú º¤ÅÍÈ­µÈ ¹®¼­¸¦ ºÐ·ùÇÏ´Â °úÁ¤ À¸·Î ³ª´­ ¼ö ÀÖ´Ù. ±âÁ¸ ¿¬±¸µéÀÇ ¹æ½ÄµéÀ» ¸®ºäÇÏ°í ¾î¶² ¹æ½ÄÀÇ ÀÓº£µù ¹æ¹ý°ú µö·¯´× ¸ðµ¨ÀÌ Çѱ¹¾î ¹®¼­¿¡ ÀûÇÕÇÑÁö °¨¼ººÐ¼®¿¡ ºñ±³ ½ÇÇèÀ» ÅëÇØ Çѱ¹¾î¿¡ ÀûÇÕÇÑ ¹æ¹ý·ÐÀ» ã¾Æ³½´Ù. ¹®¼­ Àüó¸® ¹æ¹ýÀº ¹®¼­¸¦ ´Ü¾î, À½Àý ±×¸®°í À½¼Ò ´ÜÀ§·Î ÅäÅ«È­ ÇÏ´Â ¹æ¹ýÀ» ºñ±³ÇÏ¿´´Ù. ¶ÇÇÑ, ¸ðµ¨À» CNNºÎÅÍ LSTM, Bi-LSTM, Stacked Bi-LSTM, »õ·Ó°Ô Á¦¾ÈÇÏ´Â Parallel Stacked Bidirectional LSTM ¸ðµ¨±îÁö ³×À̹ö ¿µÈ­ ¸®ºä µ¥ÀÌÅͼÂÀÎ NSMC¿¡ ´ëÇØ ºñ±³ ½ÇÇèÀ» ÇÏ¿´´Ù. Á¦¾ÈµÈ ¸ðµ¨ÀÇ ¼º´ÉÀÌ ±âÁ¸ÀÇ ±âº» µö·¯´× ¸ð µ¨¿¡ ºñÇØ ³ôÀº ¼º´ÉÀ» º¸ÀÓÀ» È®ÀÎÇÏ¿´°í, ´Ù¸¥ Àü󸮸¦ ÅëÇØ ÇнÀµÈ ¸ðµ¨°£ÀÇ ¾Ó»óºíÀ» ÅëÇØ º¸´Ù ÃÖ °í ¼º´ÉÀÎ 88.95%ÀÇ ºÐ·ù Á¤È®µµ¸¦ ´Þ¼ºÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
The sentiment analysis is a field of document classification that classifies the sensitivity of text documents. The sentiment analysis methodology that employs the use of deep learning can be divided into a process of tokenizing a document, obtaining a sentence vector through embedding and classifying a vectorized document. We reviewed the methods of various existing studies and found out the appropriate methodology focusing on embedding methods and deep learning models for the Korean documents through comparative experiments. The document pre-processing method compares documents to words, syllables and phonemes. Additionally, a comparative experiment was conducted on the Naver movie review data set nsmc (naver sentiment movie corpus) from the CNN to the LSTM, bi-LSTM, stacked bi-LSTM and the newly proposed Parallel Stacked Bidirectional LSTM model. The results showed that the performance of the proposed model was higher than that of the existing simple deep learning model. Moreover, itachieved the best classification performance of approximately 88.95% through the ensemble among the models learned through other pre-processing.
Å°¿öµå(Keyword) °¨¼º ºÐ¼®   Çѱ¹¾î ÀÚ¿¬¾î 󸮠  recurrent neural networks   word2vec   long short term memory   sentiment analysis   korean natural language proces  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå