• 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 : 16 / 16

ÇѱÛÁ¦¸ñ(Korean Title) °¨¿°º´ÀÇ ¿¹Ãø ½ÃÁ¡¿¡ µû¸¥ º¯¼ö Áß¿äµµ º¯È­ ¿¬±¸
¿µ¹®Á¦¸ñ(English Title) A study on the importance change of variables according to the prediction point of infectious diseases
ÀúÀÚ(Author) Á¤½Â¿ø   ¹®Àç¿í   ½ÉÁ¾È­   ȲÀÎÁØ   Seungwon Jung   Jaeuk Moon   Jonghwa Shim   Eenjun Hwang  
¿ø¹®¼ö·Ïó(Citation) VOL 36 NO. 03 PP. 0020 ~ 0035 (2020. 12)
Çѱ۳»¿ë
(Korean Abstract)
°¨¿°º´ÀÇ È¿°úÀû ´ëÀÀÀ» À§Çؼ­´Â, °¨¿°º´ ȯÀÚ ¼ö¿¡ ´ëÇÑ Á¤È®ÇÑ »çÀü ¿¹ÃøÀÌ Áß¿äÇÏ´Ù. À̸¦ À§ÇØ, °¨ ¿°º´ÀÇ Æ¯¼ºÀ» °í·ÁÇÑ ¼öÇÐÀû ¸ðµ¨¸µ Á¢±Ù¹ýÀÌ ÁÖ·ù¸¦ ÀÌ·ç°í ÀÖÀ¸³ª, °¨¿°º´ÀÇ Æ¯¼º¿¡ °üÇÑ »çÀü Áö½ÄÀÌ ¿ä±¸µÇ´Â ÇÑ°èÁ¡ÀÌ ÀÖ´Ù. ÃÖ±Ù¿¡´Â ±â»óÀ̳ª °ú°Å ¹ß»ý µ¥ÀÌÅÍ¿Í °°Àº °¨¿° °ü·Ã ÀڷḦ È°¿ëÇÑ ±â°èÇнÀ ±â¹Ý ¿¹Ãø ¸ðµ¨ÀÌ ÁÖ¸ñÀ» ¹Þ°í ÀÖ´Ù. ÀÌ ¹æ¹ýÀº ±âÁ¸ ¹æ½Äº¸´Ù Á¦¾àÀÌ ´úÇϳª, ³ôÀº ¿¹Ãø Á¤È®µµ¸¦ À§Çؼ­´Â ÀÔ·Â º¯¼ö·Î »ç¿ëÇÒ ¿ä¼Òµé¿¡ ´ëÇÑ ½ÅÁßÇÑ ¼±ÅÃÀÌ ¿ä±¸µÈ´Ù. ÇÏÁö¸¸, °¢ ¿ä¼ÒÀÇ Á߿伺Àº ¿¹Ãø ½ÃÁ¡¿¡ µû¶ó ´Þ¶óÁú ¼ö ÀÖ¾î, ¼±ÅÃÀÌ ½±Áö ¾ÊÀ¸¸ç, À̸¦ ÇØ°áÇϱâ À§ÇÑ ¿¬±¸ ¶ÇÇÑ ºÎÁ·ÇÑ ½ÇÁ¤ÀÌ´Ù. ÀÌ¿¡ º» ³í¹®¿¡¼­´Â ·£´ý Æ÷·¹½ºÆ® ±â¹Ý °¨¿°º´ ¿¹Ãø ¸ðµ¨À» ±¸¼ºÇÏ¿©, ¿¹Ãø ½ÃÁ¡ º¯È­¿¡ µû¸¥ º¯¼ö Áß¿äµµ º¯È­¸¦ ºÐ¼®ÇÏ´Â ¿¬±¸ ¸¦ ¼öÇàÇÑ´Ù. ¸ÕÀú, °¨¿°º´ ¹ß»ý Á¤º¸, ±â»ó Á¤º¸ µî °ü·Ã ÀڷḦ ¼öÁýÇÏ°í Àüó¸®ÇÏ¿© µ¥ÀÌÅͼÂÀ» ±¸¼ºÇÑ´Ù. ±¸¼ºµÈ µ¥ÀÌÅͼÂÀ» ±â¹ÝÀ¸·Î ¿¹Ãø ½ÃÁ¡À» º¯È­½ÃÅ°¸é¼­ ·£´ý Æ÷·¹½ºÆ®¸¦ ÇнÀ½ÃÅ°°í º¯¼ö Áß¿äµµ¸¦ ºÐ¼®ÇÑ ´Ù. ±¹³» ¹ß»ý °¨¿°º´À» ´ë»óÀ¸·Î ½ÇÇèÀ» ¼öÇàÇÑ °á°ú, º¯¼ö Áß¿äµµ º¯È­ ¾ç»ó°ú °¨¿°º´ÀÇ ¹ß»ý ÆÐÅÏ¿¡ µû¶ó °¨¿°º´À» ¼¼ °¡Áö À¯ÇüÀ¸·Î ºÐ·ùÇÒ ¼ö ÀÖ¾úÀ¸¸ç, À¯Çü¸¶´Ù Áß¿ä ÀÔ·Â º¯¼ö, ÁÖ¿ä º¯È­ ¾ç»ó µîÀ» ºÐ¼®ÇÒ ¼ö ÀÖ¾ú´Ù
¿µ¹®³»¿ë
(English Abstract)
In order to cope with infectious diseases, it is important to accurately predict the number of patients who will suffer from them. The traditional approach for this was to construct a mathematical model that considers the characteristics of infectious diseases, so prior knowledge about their characteristics is required. Recently, machine learning-based approach based on relevant data such as weather and past occurrences has drawn much attention. Although this approach is free from the limitation of the traditional approach, the input variables should be selected so that good prediction performance can be achieved. However, the importance of each input changes depending on when the prediction point is. In this paper, we propose a random forest-based prediction model for analyzing the change of variable importances according to the prediction point. To do this, we first collect and pre-process relevant data for the analysis. Then, we train a random forest and obtain the variable importances while changing the prediction point. To evaluate the effectiveness of our approach, we conducted various experiments for domestic infectious diseases and classified them into three clusters. Then, we extracted features from each cluster, such as important input variables and major changes.
Å°¿öµå(Keyword) °¨¿°º´ ¿¹Ãø   ·£´ý Æ÷·¹½ºÆ®   º¯¼ö Áß¿äµµ   ±â°è ÇнÀ   Infectious disease forecasting   random forest   feature importance   machine learning  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå