• 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 : 475 / 476

ÇѱÛÁ¦¸ñ(Korean Title) ¹®¼­ ½Ö À¯»çµµ ÆǺ°À» À§ÇÑ ¹®Àå »óÈ£ °ü°è ¹× ±×·¡ÇÁ ±â¹Ý ¸ðµ¨ÀÇ ¾Ó»óºí
¿µ¹®Á¦¸ñ(English Title) Ensemble of Sentence Interaction and Graph Based Models for Document Pair Similarity Estimation
ÀúÀÚ(Author) ÃÖ¼ºÈ¯   ¼Õµ¿Çö   ÀÌȣâ   Seonghwan Choi   Donghyun Son   Hochang Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 48 NO. 11 PP. 1184 ~ 1193 (2021. 11)
Çѱ۳»¿ë
(Korean Abstract)
´º½º ±â»ç¿Í °°Àº ¹®¼­ Ŭ·¯½ºÅ͸µ¿¡¼­ µÎ ¹®¼­ °£ÀÇ À¯»çµµ´Â Ŭ·¯½ºÅÍÀÇ Æ¯¼ºÀ» °áÁ¤ÇÏ´Â Áß¿äÇÑ ºÎºÐ Áß ÇϳªÀÌ´Ù. ±âÁ¸ µö·¯´× ±â¹Ý Á¢±Ù ¹æ¹ýÀÎ ½ÃÄö½º À¯»çµµ ÃøÁ¤ ¸ðµ¨Àº ¹®¼­ ´ÜÀ§¿¡¼­ ³ªÅ¸³ª´Â ±ä ¹®¸ÆÀ» ¹Ý¿µÇÏÁö ¸øÇÏ´Â ¹®Á¦Á¡À» °¡Áö°í ÀÖ´Ù. º» ¿¬±¸¿¡¼­´Â ´º½º Ŭ·¯½ºÅ͸µ¿¡ ÀûÇÕÇÑ ¹®¼­½Ö À¯»çµµ ¸ðµ¨À» ±¸¼ºÇϱâ À§ÇØ »óÈ£ ÀÛ¿ë ±â¹Ý Á¢±Ù, ±×·¡ÇÁ ±â¹Ý Á¢±Ù ¹æ¹ýÀ» »ç¿ëÇÑ´Ù. »óÈ£ ÀÛ¿ë ±â¹Ý Á¢±Ù¿¡¼­´Â ¹®¼­ ½Ö ³» ´Ù¼öÀÇ ¹®Àå Ç¥Çöµé °£ÀÇ À¯»çµµ Á¤º¸¸¦ Á¾ÇÕÇØ Àüü ¹®¼­ ½ÖÀÇ À¯»çµµ¸¦ ÃøÁ¤ÇÏ´Â ³× °¡Áö À¯»çµµ ¸ðµ¨À» Á¦¾ÈÇÑ´Ù. ±âÁ¸ Á¢±Ù ¹æ¹ýµéÀÎ SVM, HAN¿¡ ºñÇØ µÎ °¡Áö Á¢±Ù ¹æ¹ý¿¡¼­ ³ôÀº ¼º´ÉÀÌ ³ªÅ¸³²À» È®ÀÎÇß´Ù. ±×·¡ÇÁ ±â¹Ý Á¢±Ù¿¡¼­´Â ÀԷ¿¡ »ç¿ëµÇ´Â ÀÚÁúÀÇ Á¾·ù¿Í ½Å°æ¸ÁÀÇ ±íÀÌ¿¡ µû¸¥ ¼º´É º¯È­¸¦ È®ÀÎÇß´Ù. ¶ÇÇÑ, »óÀÌÇÑ µÎ Á¢±Ù ¹æ¹ýÀÌ °®´Â ¿¹Ãø¾ç»óÀÇ Â÷ÀÌ¿Í »óÈ£º¸¿Ï¼ºÀ» ¿À·ù ºÐ¼®°ú ¾Ó»óºíÀ» ÅëÇØ È®ÀÎÇß´Ù.
¿µ¹®³»¿ë
(English Abstract)
Deriving the similarity between two documents, such as, news articles, is one of the most important factors of clustering documents. Sequence similarity models, one of the existing deep-learning based approaches to document clustering, do not reflect the entire context of documents. To address this issue, this paper uses interaction-based and graph-based approaches to construct document pair similarity models suitable for news clustering. This paper proposes four interaction-based models that measures the similarity between two documents through the aggregation of similarity information in the interaction of sentences. The experimental results demonstrated that two out of these four proposed models outperformed SVM and HAN. Ablation studies were conducted on the graph-based model through experiments on the depth of the model¡¯s neural network and its input features. Through error analysis and ensemble of models with an interaction and graph-based approach, this paper showed that these two approaches could be complementarity due to the differences in their prediction tendencies.
Å°¿öµå(Keyword) ´º½º Ŭ·¯½ºÅ͸µ   ¹®¼­ ½Ö À¯»çµµ   ÅؽºÆ® À¯»çµµ   »óÈ£ ÀÛ¿ë ±â¹Ý   ±×·¡ÇÁ ±â¹Ý   news clustering   document similarity   text similarity   interaction-based   graph-based  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå