• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ÇÐȸÁö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ÇÐȸÁö > µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) °³ÀÎ ¿îÀüÀÚÀÇ ÁÖÇà Ư¼ºÀ» ¹Ý¿µÇÑ ¼Óµµ ¿¹Ãø ±â¹ý
¿µ¹®Á¦¸ñ(English Title) °³ÀÎ ¿îÀüÀÚÀÇ ÁÖÇà Ư¼ºÀ» ¹Ý¿µÇÑ ¼Óµµ ¿¹Ãø ±â¹ý
ÀúÀÚ(Author) Á¶¼®Çö   ÃÖ°æÀÏ   À̼®·æ   Seok-Hyun Cho   Kyung-Il Choe   Seok-Lyong Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 33 NO. 01 PP. 0048 ~ 0061 (2017. 04)
Çѱ۳»¿ë
(Korean Abstract)
Áö´ÉÇü ±³Åëü°è (ITS: intelligent transport system)ÀÇ È®»ê°ú ±³ÅëÁ¤º¸ ÀúÀå ±â¼úÀÇ ¹ß´Þ·Î ÀÎÇÏ¿© Àü±¹ÀÇ µµ·Î¿¡¼­ ÇÏ·ç¿¡µµ ¼öõ¸¸ °ÇÀÇ ±³Åë ¼Óµµ µ¥ÀÌÅ͵éÀÌ ¼öÁýµÇ°í ÀÖÀ¸¸ç, ½Ã°£ÀÇ È帧¿¡ µû¶ó »ý¼ºµÇ´Â ¹æ´ëÇÑ ¾çÀÇ µ¥ÀÌÅ͸¦ È¿À²ÀûÀ¸·Î ó¸®ÇÏ´Â ¹æ¹ý°ú ½Å·Ú¼º ÀÖ´Â ÅëÇà ¼Óµµ Á¤º¸ ¿¹ÃøÀÌ Áß¿äÇÑ ¿¬±¸ ÁÖÁ¦°¡ µÇ°í ÀÖ´Ù. ÀÌ·¯ÇÑ ±³Åë Á¤º¸ ½Ã½ºÅÛÀÇ ¹ß´Þ·Î ±³Åë Á¤º¸¿¡ ´ëÇÑ ´Ù¾çÇÑ ºÐ¼®ÀÌ °¡´ÉÇØ Á³Áö¸¸ ¿îÀüÀÚÀÇ °³ÀΠƯ¼ºÀ» ¹Ý¿µÇÑ ±³Åë Á¤º¸ÀÇ ºÐ¼®Àº ¾ÆÁ÷±îÁö ¿¬±¸°¡ °ÅÀÇ ÀÌ·ç¾îÁö°í ÀÖÁö ¾ÊÀº ºÐ¾ßÀÌ´Ù. º» ³í¹®¿¡¼­´Â µµ·Î À¯Çü, ½Ã°£´ë µîÀÇ ¼Ó¼º (attribute)À» °í·ÁÇÏ¿© ¿îÀüÀÚÀÇ °³ÀΠƯ¼ºÀ» ¹Ý¿µÇÑ ¼Óµµ ¿¹Ãø ±â¹ýÀ» Á¦½ÃÇÑ´Ù. µ¥ÀÌÅÍ Á¤±ÔÈ­¸¦ ÅëÇÏ¿© Àüü ºÐÆ÷¿¡¼­ »ó´ëÀû À§Ä¡¸¦ ³ªÅ¸³»´Â Åë°è·®ÀÎ z-score¸¦ ÀÌ¿ëÇÏ¿© ¿îÀüÀÚÀÇ °³ÀÎ ÁÖÇà Ư¼ºÀ» Á¤ÀÇÇÏ°í, °³ÀΠƯ¼º¿¡ µû¶ó ¿îÀüÀÚµéÀ» ±ºÁý ±â¹ýÀÎ k-means clustering ¾Ë°í¸®ÁòÀ» »ç¿ëÇÏ¿© ±ºÁýÈ­ ÇÔÀ¸·Î½á ±ºÁý ´ÜÀ§ÀÇ ¼Óµµ Á¤º¸ °ü¸® ¹æ¾ÈÀ» Á¦¾ÈÇÑ´Ù. ´ÙÀ½À¸·Î, ±ºÁýº°·Î °è»êµÈ ¿îÀüÀÚ ÁÖÇà Ư¼ºÀ» ¹Ý¿µÇÏ¿© ¸ñÀûÁö µµ´Þ ½Ã°£ (destination arrival time)À» °è»êÇϱâ À§ÇÑ ¼Óµµ¸¦ º¸Á¤ÇÔÀ¸·Î½á ¼Óµµ ¿¹Ãø¿¡ ´ëÇÑ Á¤È®µµ¸¦ °³¼±ÇÑ´Ù. ½ÇÇè °á°ú, º» ¿¬±¸¿¡¼­ Á¦½ÃÇÑ ¿¹Ãø ±â¹ýÀ» »ç¿ëÇÑ °æ¿ì, Æò±Õ 5.27%, ÃÖ´ë 10.4%ÀÇ ¿ÀÂ÷ °³¼± È¿°ú¸¦ ´Þ¼ºÇÒ ¼ö ÀÖ¾ú´Ù.
¿µ¹®³»¿ë
(English Abstract)

With the development of ITS (intelligent transport system) and traffic information storage technology, tens of millions of data are gathered nationwide on daily basis. It is becoming an important research issue recently to handle such large amount of data efficiently and to perform the reliable traffic speed prediction. It has become possible to analyze various traffic information with the development of the intelligent traffic system, but data analysis that reflects driver's individual driving characteristics has not been considered yet. In this study, we present the speed prediction method considering various attributes such as road types and daily time slots. We have defined a driver's driving characteristic using z-score, which is a statistics for ascertaining the relative position in the entire distribution by normalizing the distribution. We have also presented a process of managing the speed data according to the groups that are generated by clustering drivers using the k-means clustering algorithm. Finally, we propose a new prediction method that corrects the speed for calculating destination arrival time by reflecting the driver¡¯s driving characteristic. Result analysis shows a noteworthy effect by the improvement of error prediction, 5.27% on average and up to 10.4%, using the correction method of speed prediction proposed in this study.
Å°¿öµå(Keyword) Áö´ÉÇü ±³Åë ü°è   ¼Óµµ ¿¹Ãø   °³ÀÎÈ­   µ¥ÀÌÅ͸¶ÀÌ´×   ±ºÁýÈ­È­   intelligent transport system   speed prediction   personalization   data mining   clustering  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå