• 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¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Current Result Document : 12 / 14

ÇѱÛÁ¦¸ñ(Korean Title) ¼±¹Ú µ¥ÀÌÅÍ¿¡¼­ ¿¹Ãø Á¤È®¼º Çâ»óÀ» À§ÇÑ ÈÆ·Ã µ¥ÀÌÅÍÀÇ ÀÌ»óÄ¡ °ËÃâ ¹æ¾È
¿µ¹®Á¦¸ñ(English Title) Detecting Outliers on Training Dataset for Better Quality of Estimation on Vessel Traces
ÀúÀÚ(Author) Á¤Çü±Ù   ÀÌ°­¿ì   Á¶Àº¼±   Hyungkun Jung   Kang-Woo Lee   Eun-Sun Cho  
¿ø¹®¼ö·Ïó(Citation) VOL 25 NO. 12 PP. 0594 ~ 0601 (2019. 12)
Çѱ۳»¿ë
(Korean Abstract)
ÃÖ±Ù ÇØ¾ç ¼ö¼Û¾÷ÀÇ ¾ÈÀü¼ºÀ» ³ôÀ̱â À§ÇÏ¿© ¼±¹Ú ÇØ¾ç µ¥ÀÌÅÍ ºÐ¼®ÀÇ Çʿ伺ÀÌ ´ëµÎµÇ°í ÀÖ´Ù. ÀÌ·¯ÇÑ ºÐ¼®Àº ¼±¹ÚÀÇ µµÂø ½Ã°£À̳ª µµÂø Ç×±¸ ¿¹Ãø µîÀ» ±× ¸ñÀûÀ¸·Î ÇÑ´Ù. º» ³í¹®¿¡¼­´Â EU2020ÀÇ Marin Traffic»ç¿Í BigDataOcean ÇÁ·ÎÁ§Æ®¿¡¼­ Á¦°øÇÏ´Â ½ÇÁ¦ Ç×ÇØ µ¥ÀÌÅ͸¦ ¹ÙÅÁÀ¸·Î µµÂø ½Ã°£ ¹× Ç×±¸¸¦ ¿¹ÃøÇÏ´Â ¹æ¹ý¿¡ ´ëÇØ ¼Ò°³ÇÏ°í, Á¤È®¼ºÀ» ³ôÀ̱â À§ÇÑ ¹æ¾ÈÀ¸·Î ½ÇÁ¦ ÈÆ·Ã µ¥ÀÌÅÍ¿¡ Æ÷ÇÔµÈ ¾Æ¿ô ¶óÀ̾ ŽÁöÇÏ´Â ¹æ¾È¿¡ ´ëÇØ Á¦¾ÈÇÑ´Ù. ±ËÀûÀÇ ¾Æ¿ô¶óÀ̾î ŽÁö¸¦ À§ÇÏ¿© ±âÁ¸ÀÇ ±ºÁýÈ­ ±â¹ýÀ» »ç¿ëÇÏ¿© Ãß°¡·Î ´ëÇ¥ ±ËÀûÀ» ¼±ÃâÇÏ¿© Á¤Â÷ ¼±¹ÚÀ» ŽÁöÇÏ°í, ¾Æ¿ô¶óÀÌ¾î ±ËÀû °ª 󸮿¡ ´ëÇؼ­ ¼³¸íÇÑ´Ù. º» ³í¹®¿¡¼­ Á¦¾ÈÇÏ´Â ¹æ¹ýµéÀÇ ½Ç¿ë¼ºÀ» º¸À̱â À§ÇÏ¿© ¾Æ¿ô¶óÀ̾ Á¦°ÅÇÑ ¼±¹ÚÀÇ µµÂø ½Ã°£ ¹× Ç×±¸ ¿¹ÃøÀÌ ±âÁ¸ ¹æ¹ý¿¡ ºñÇØ Çâ»óµÊÀ» º¸¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
Recently, data analysis methods have emerged to improve the safety of the maritime transport industry. The focus of these analyses frequently include estimation of arrival ports and arrival times. This paper introduces an estimation method for arrival ports and arrival times from ship trajectory data based on actual maritime data provided by Marin and the BigDataOcean Project of EU2020, and proposes a new outlier detection method from training data set, to enhance the quality of the estimation method. We use clustering techniques in existing studies to detect trajectory outliers. Our study focuses on selecting representative trajectories to detect parked vessels, and how to manipulate trajectory outlier values. To demonstrate the practicality of the proposed method, we apply the algorithm to the model of the existing study, which shows better quality of estimation on the destinations and the arrival time of the ships.
Å°¿öµå(Keyword) ¼±¹Ú ±ËÀû µ¥ÀÌÅÍ   ¾Æ¿ô¶óÀ̾ ±ºÁýÈ­   ÇϿ콺µµ¸£ÇÁ °Å¸®   vessel trajectory data   outlier   clustering   Hausdorff distance  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå