• 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 : 1 / 2

ÇѱÛÁ¦¸ñ(Korean Title) »çȸÀû º¯¼ö¸¦ °í·ÁÇÑ LSTM ±â¹Ý Äڷγª19 ÀϺ° È®ÁøÀÚ ¼ö ¿¹Ãø ±â¹ý
¿µ¹®Á¦¸ñ(English Title) LSTM-based Daily COVID-19 Forecasting Scheme Considering Social Variables
ÀúÀÚ(Author) ÀÓ½ÂÈ£   ¹ÚÂùÇö   Seungho Lim   Chanhyun Park   ³ëÀ±¾Æ   Á¤½Â¿ø   ¹®Àç¿í   ȲÀÎÁØ   Yoona Noh   Seungwon Jung   Jaeuk Moon   Eenjun Hwang  
¿ø¹®¼ö·Ïó(Citation) VOL 28 NO. 02 PP. 0116 ~ 0121 (2022. 02)
Çѱ۳»¿ë
(Korean Abstract)
Äڷγª19ÀÇ È®»êÀ¸·Î ÀÎÇØ ¼¼°èÀûÀ¸·Î ¸¹Àº ÇÇÇØ°¡ º¸°íµÇ°í ÀÖ´Ù. ÇÇÇظ¦ ÁÙÀ̱â À§Çؼ­´Â ¹ÙÀÌ·¯½ºÀÇ È®»êÀ» Á¶±â¿¡ ÆľÇÇÏ°í ÀûÀýÇÑ ´ëÃ¥À» ½Å¼ÓÈ÷ ÁغñÇÏ´Â °ÍÀÌ Áß¿äÇÏ´Ù. À̸¦ À§ÇØ ½Ã°è¿­ ¿¹Ãø¿¡ Ź¿ùÇÑ LSTM ¸ðµ¨°ú °°Àº ±â°èÇнÀ ¸ðµ¨À» È°¿ëÇÑ Äڷγª19 È®ÁøÀÚ ¼ö ¿¹Ãø ¿¬±¸°¡ ¼öÇàµÇ¾úÀ¸³ª, ´ëºÎºÐ È®ÁøÀÚ ¼ö µ¥ÀÌÅ͸¸ ÀÔ·Â º¯¼ö·Î »ç¿ëÇÏ¿© ´Ù¼Ò ºÎÁ¤È®Çß´Ù. ÇÑÆí, Äڷγª19 ´ëÀÀ Á¤Ã¥À̳ª À¯µ¿Àα¸ µîÀº »çȸÀû º¯È­¸¦ ³ªÅ¸³»±â¿¡ ¿¹Ãø ¸ðµ¨¿¡ Àû¿ëµÉ ÇÊ¿ä°¡ ÀÖ´Ù. ƯÈ÷, ¹é½Å Á¢Á¾À¸·Î ÀÎÇØ º¯È­µÈ È®»ê ÆÐÅÏÀ» ¿¹Ãø °¡´ÉÇÑ ¸ðµ¨ÀÌ ÇÊ¿äÇÏ´Ù. ÀÌ¿¡, º» ³í¹®¿¡¼­´Â »çȸÀû º¯È­¸¦ ¹Ý¿µÇÏ´Â º¯¼öµéÀ» È°¿ëÇÑ LSTM ±â¹Ý Äڷγª19 È®ÁøÀÚ ¼ö ¿¹Ãø ±â¹ýÀ» Á¦¾ÈÇÑ´Ù. ¿¹Ãø ¸ðµ¨ ±¸¼ºÀ» À§ÇØ »çȸÀû °Å¸®µÎ±â ´Ü°è, ÁöÇÏö ½ÂÂ÷ Àοø, ¡°Äڷγª¡± Å°¿öµå °Ë»ö·®, ¹é½Å Á¢Á¾ÀÚ ¼ö µîÀÇ µ¥ÀÌÅ͸¦ ¼öÁýÇÏ°í ¸ðµ¨ÀÇ ÀÔ·Â º¯¼ö·Î È°¿ëÇÑ´Ù. ´Ù¾çÇÑ ¸ðµ¨°úÀÇ ºñ±³ ½ÇÇèÀ» ÅëÇÏ¿© Á¦¾ÈÇÑ ±â¹ýÀÇ ¿ì¼ö¼ºÀ» º¸ÀδÙ.
¿µ¹®³»¿ë
(English Abstract)
Due to the spread of COVID-19, many losses are being reported worldwide. In order to reduce losses, it is important to identify the spread of the virus early and prepare appropriate countermeasures quickly. Various studies have been conducted to forecast the number of COVID-19 cases using machine learning models such as the LSTM model, which shows excellent performance in time series forecasting. However, most of them do not have high forecasting accuracy because they use only confirmed cases data as input variables. Meanwhile, COVID-19 response policy, floating population, and so on that can represent social changes need to be applied to a forecasting model. In particular, due to the introduction of vaccines, a forecasting model suitable for changed spread pattern is needed. Therefore, we proposed an LSTM-based forecasting scheme to predict the number of COVID-19 confirmed cases using social variables in this study. To construct the forecasting model, we collected data such as social distancing level, number of subway passengers, "COVID-19" keyword searches, and number of vaccinations and used them as input variables. Through comparative experiments with various models, the proposed scheme demonstrated an excellent forecasting performance.
Å°¿öµå(Keyword) SSD   NVMe   ½ºÅ丮Áö ½Ã½ºÅÛ   ÄÄÇ»ÅÍ ½Ã½ºÅÛ   ¼º´É ºÐ¼®   storage system   computer system   performance analysis   Äڷγª19   °¨¿°º´ ¿¹Ãø   ¹é½Å Á¢Á¾   LSTM   COVID-19   infectious disease forecasting   vaccination   LSTM  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå