• 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) Self-AttentionÀ» È°¿ëÇÑ Siamese CNN-Bidirectional LSTM ±â¹Ý ¹®Àå À¯»çµµ ¿¹Ãø
¿µ¹®Á¦¸ñ(English Title) Sentence Similarity Prediction based on Siamese CNN-Bidirectional LSTM with Self-attention
ÀúÀÚ(Author) ±è¹ÎÅ   ¿À¿µÅà  ±è¿ìÁÖ   Mintae Kim   Yeongtaek Oh   Wooju Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 46 NO. 03 PP. 0241 ~ 0245 (2019. 03)
Çѱ۳»¿ë
(Korean Abstract)
º» ³í¹®¿¡¼­´Â ÀÔ·ÂµÈ µÎ ¹®ÀåÀÇ À¯»çµµ¸¦ ÃøÁ¤ÇÏ´Â µö·¯´× ¸ðµ¨À» Á¦¾ÈÇÑ´Ù. ±âÁ¸ÀÇ ¹®ÀåÀÇ À¯»çµµ ÃøÁ¤ ¸ðµ¨¿¡´Â ´Ü¾î ȤÀº ÇüÅÂ¼Ò ´ÜÀ§·Î ¹®ÀåÀ» ºÐÇØÇÏ¿© ÀÓº£µù ÇÏ´Â ¹æ½ÄÀ» È°¿ëÇÑ´Ù. ÇÏÁö¸¸ ÀÌ´Â »çÀüÀÇ Å©±â¸¦ Áõ°¡½ÃÄÑ ¸ðµ¨ÀÇ º¹Àâµµ¸¦ ³ôÀÌ´Â ¹®Á¦Á¡ÀÌ ÀÖ´Ù. º» ³í¹®¿¡¼­´Â ¹®ÀåÀ» À½¼Ò ´ÜÀ§·Î ºÐÇØÇÏ¿© ¸ðµ¨ º¹Àâµµ¸¦ ÁÙÀÌ°í ÇØ´ç À½¼Ò¸¦ ¹­¾îÁÖ´Â ´Ù¾çÇÑ ÇÊÅÍ »çÀÌÁîÀÇ 1D Convolution Neural Network¿Í Long Short Term Memory(LSTM)À» °áÇÕÇÑ Siamese CNN-Bidirectional LSTM ¸ðµ¨À» Á¦¾ÈÇÑ´Ù. º» ¸ðµ¨À» Æò°¡Çϱâ À§ÇØ ³×À̹ö Áö½ÄÀÎ µ¥ÀÌÅ͸¦ È°¿ëÇÏ¿© ±âÁ¸ÀÇ ¹®¼­ À¯»ç ÃøÁ¤¿¡¼­ ÁÁÀº ¼º´ÉÀ» º¸ÀÌ´Â ¸ðµ¨ Manhattan LSTM(MaLSTM)°ú ºñ±³ÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
A deep learning model for semantic similarity between sentences was presented. In general, most of the models for measuring similarity word use level or morpheme level embedding. However, the attempt to apply either word use or morpheme level embedding results in higher complexity of the model due to the large size of the dictionary. To solve this problem, a Siamese CNN-Bidirectional LSTM model that utilizes phonemes instead of words or morphemes and combines long short term memory (LSTM) with 1D convolution neural networks with various window lengths that bind phonemes is proposed. For evaluation, we compared our model with Manhattan LSTM (MaLSTM) which shows good performance in measuring similarity between similar questions in the Naver Q&A dataset (similar to Kaggle Quora Question Pair).
Å°¿öµå(Keyword) ÀÚ¿¬¾î 󸮠  À¯»çµµ ÃøÁ¤   ¼¤ ³×Æ®¿öÅ©   ÇÕ¼º°ö ½Å°æ¸Á   ¼øȯ ½Å°æ¸Á   ¾îÅټǠ  natural language processing   similarity measure   siamese network   convolution neural network   recurrent neural network   attention  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå