• 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) ´º·Î ½Éº¼¸¯ ±â¹Ý ±ÔÄ¢ À¯µµ ¹× Ãß·Ð ¿£ÁøÀ» È°¿ëÇÑ Áö½Ä ¿Ï¼º ½Ã½ºÅÛ
¿µ¹®Á¦¸ñ(English Title) Knowledge Completion System using Neuro-Symbolic-based Rule Induction and Inference Engine
ÀúÀÚ(Author) ½Å¿øö   ¹ÚÇö±Ô   ¹Ú¿µÅà  Won-Chul Shin   Hyun-Kyu Park   Young-Tack Park  
¿ø¹®¼ö·Ïó(Citation) VOL 48 NO. 11 PP. 1202 ~ 1210 (2021. 11)
Çѱ۳»¿ë
(Korean Abstract)
ÃÖ±Ù Áö½Ä ±×·¡ÇÁÀÇ ºÒ¿ÏÀü¼º ¹®Á¦¸¦ ÇØ°áÇϱâ À§ÇÑ ´Ù¾çÇÑ Áö½Ä ¿Ï¼º ¿¬±¸Áß µö·¯´× ÇнÀ ¹æ¹ý°ú ·ÎÁ÷ ½Ã½ºÅÛÀÇ ÀåÁ¡À» °áÇÕÇÑ NTP(Neural Theorem Prover)¿Í °°Àº ¿¬±¸°¡ ±âÁ¸ ¿¬±¸µé¿¡ ºñÇØ ÁÁÀº ¼º´ÉÀ» ³»°í ÀÖ´Ù. ÇÏÁö¸¸ NTP´Â ÇϳªÀÇ ÀԷ¿¡ ´ëÇÑ ¿¹Ãø °á°ú¸¦ ¾ò±â À§ÇØ Áö½Ä ±×·¡ÇÁÀÇ ¸ðµç Æ®¸®ÇÃÀÌ ¿¬»ê¿¡ °ü¿©ÇÏ°Ô µÇ¹Ç·Î ´ë¿ë·® Áö½Ä ±×·¡ÇÁ 󸮿¡ ÇÑ°è°¡ ÀÖ´Ù. º» ³í¹®¿¡¼­´Â NTPÀÇ °è»ê º¹Àâµµ ¹®Á¦¸¦ °³¼±ÇÑ ¸ðµ¨·ÎºÎÅÍ ½Éº¼ÀÇ º¤ÅÍ Ç¥ÇöÀ» ÇнÀÇÏ¿© ±ÔÄ¢À» À¯µµÇÏ°í, Ãß·Ð ¿£ÁøÀ» »ç¿ëÇÏ¿© À¯µµµÈ ±ÔÄ¢À¸·ÎºÎÅÍ Áö½Ä Ãß·ÐÀ» ¼öÇàÇÒ ¼ö ÀÖ´Â µö·¯´× ÇнÀ ¹æ½Ä°ú ·ÎÁ÷ Ãß·Ð ¹æ½ÄÀÇ ÅëÇսýºÅÛÀ» Á¦¾ÈÇÑ´Ù. º» ³í¹®¿¡¼­ »ç¿ëÇÑ ±ÔÄ¢ »ý¼º¸ðµ¨ÀÇ ±ÔÄ¢À¯µµ ¼º´É °ËÁõÀ» À§ÇØ NTP¿Í Nations, Kinship, UMLS µ¥ÀÌÅÍ ¼ÂÀ» ´ë»óÀ¸·Î À¯µµµÈ ±ÔÄ¢À» È°¿ëÇÑ Å×½ºÆ® µ¥ÀÌÅÍ Ã߷а¡´É ¿©ºÎ¸¦ ºñ±³ÇÏ¿´À¸¸ç, ´ë±Ô¸ð Áö½Ä±×·¡ÇÁÀÎ Kdata¿Í WiseKB¸¦ »ç¿ëÇÑ ½ÇÇè¿¡¼­´Â Ãß·Ð ¿£ÁøÀ» ÅëÇÑ Áö½Ä Ãß·Ð °á°ú ½ÇÇè¿¡ »ç¿ëµÈ Áö½Ä ±×·¡ÇÁ¿¡ ºñÇØ °¢°¢ Kdata´Â 30%, WiseKB´Â 95%Áõ°¡µÈ Áö½Ä ±×·¡ÇÁ¸¦ ¾òÀ» ¼ö ÀÖ¾ú´Ù.
¿µ¹®³»¿ë
(English Abstract)
Recently, there have been several studies on knowledge completion methods aimed to solve the incomplete knowledge graphs problem. Methods such as Neural Theorem Prover (NTP), which combines the advantages of deep learning methods and logic systems, have performed well over existing methods. However, NTP faces challenges in processing large-scale knowledge graphs because all the triples of the knowledge graph are involved in the computation to obtain prediction results for one input. In this paper, we propose an integrated system of deep learning and logic inference methods that can learn vector representations of symbols from improved models of computational complexity of NTP to rule induction, and perform knowledge inference from induced rules using inference engines. In this paper, for rule-induction performance verification of the rule generation model, we compared test data inference ability with NTP using induced rules on Nations, Kinship, and UMLS data set. Experiments with Kdata and WiseKB knowledge inference through inference engines resulted in a 30% increase in Kdata and a 95% increase in WiseKB compared to the knowledge graphs used in experiments.
Å°¿öµå(Keyword) Áö½Ä ±×·¡ÇÁ   µö·¯´×   ·ÎÁ÷ ½Ã½ºÅÛ   Ãß·Ð ¿£Áø   Áö½Ä Ã߷Р  knowledge graphs   deep learning   logic system   inference engine   knowledge inference  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå