Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
Temporal Fusion Transformers¿Í ½ÉÃþ ÇнÀ ¹æ¹ýÀ» »ç¿ëÇÑ ´ÙÃþ ¼öÆò ½Ã°è¿ µ¥ÀÌÅÍ ºÐ¼® |
¿µ¹®Á¦¸ñ(English Title) |
Temporal Fusion Transformers and Deep Learning Methods for Multi-Horizon Time Series Forecasting |
ÀúÀÚ(Author) |
±èÀΰæ
±è´ëÈñ
ÀÌÀ籸
InKyung Kim
DaeHee Kim
Jaekoo Lee
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 11 NO. 02 PP. 0081 ~ 0086 (2022. 02) |
Çѱ۳»¿ë (Korean Abstract) |
½Ã°è¿ µ¥ÀÌÅÍ´Â ÁÖ½Ä, IoT, °øÀå ÀÚµ¿È¿Í °°Àº ´Ù¾çÇÑ ½Ç»ýÈ°¿¡¼ ¼öÁýµÇ°í È°¿ëµÇ°í ÀÖÀ¸¸ç, Á¤È®ÇÑ ½Ã°è¿ ¿¹ÃøÀº ÇØ´ç ºÐ¾ß¿¡¼ ¿î¿µ È¿À²¼ºÀ» ³ôÀÏ ¼ö ÀÖ¾î¼ ÀüÅëÀûÀ¸·Î Áß¿äÇÑ ¿¬±¸ ÁÖÁ¦ÀÌ´Ù. Àü¹ÝÀûÀÎ ½Ã°è¿ µ¥ÀÌÅÍÀÇ Çâ»óµÈ Ư¡À» ÃßÃâÇÒ ¼ö ÀÖ´Â ´ëÇ¥ÀûÀÎ ½Ã°è¿ µ¥ÀÌÅÍ ºÐ¼® ¹æ¹ýÀÎ ´ÙÃþ ¼öÆò ¿¹ÃøÀº ÃÖ±Ù ºÎ°¡Àû Á¤º¸¸¦ Æ÷ÇÔÇÏ´Â ½Ã°è¿ µ¥ÀÌÅÍ¿¡ ³»ÀçÇÑ ÀÌÁú¼º(heterogeneity)±îÁö Æ÷°ýÀûÀ¸·Î ºÐ¼®¿¡ È°¿ëÇÏ¿© Çâ»óµÈ ½Ã°è¿ ¿¹ÃøÇÑ´Ù. ÇÏÁö¸¸ ´ëºÎºÐÀÇ ½ÉÃþ ÇнÀ ±â¹Ý ½Ã°è¿ ºÐ¼® ¸ðµ¨µéÀº ½Ã°è¿ µ¥ÀÌÅÍÀÇ ÀÌÁú¼ºÀ» ¹Ý¿µÇÏÁö ¸øÇß´Ù. µû¶ó¼ ¿ì¸®´Â Àß ¾Ë·ÁÁø temporal fusion transformers ¹æ¹ýÀ» »ç¿ëÇÏ¿© ½Ç»ýÈ°°ú ¹ÐÁ¢ÇÑ ½ÇÁ¦ µ¥ÀÌÅ͸¦ ÀÌÁú¼ºÀ» °í·ÁÇÑ ´ÙÃþ ¼öÆò ¿¹Ãø¿¡ Àû¿ëÇÏ¿´´Ù. °á°úÀûÀ¸·Î ÁÖ½Ä, ¹Ì¼¼¸ÕÁö, Àü±â ¼Òºñ·®°ú °°Àº ½Ç»ýÈ° ½Ã°è¿ µ¥ÀÌÅÍ¿¡ Àû¿ëÇÑ ¹æ¹ýÀÌ ±âÁ¸ ¿¹Ãø ¸ðµ¨º¸´Ù Çâ»óµÈ Á¤È®µµ¸¦ °¡ÁüÀ» È®ÀÎÇÒ ¼ö ÀÖ¾ú´Ù. |
¿µ¹®³»¿ë (English Abstract) |
Given that time series are used in various fields, such as finance, IoT, and manufacturing, data analytical methods for accurate time-series forecasting can serve to increase operational efficiency. Among time-series analysis methods, multi-horizon forecasting provides a better understanding of data because it can extract meaningful statistics and other characteristics of the entire time-series. Furthermore, time-series data with exogenous information can be accurately predicted by using multi-horizon forecasting methods. However, traditional deep learning-based models for time-series do not account for the heterogeneity of inputs. We proposed an improved time-series predicting method, called the temporal fusion transformer method, which combines multi-horizon forecasting with interpretable insights into temporal dynamics. Various real-world data such as stock prices, fine dust concentrates and electricity consumption were considered in experiments. Experimental results showed that our temporal fusion transformer method has better time-series forecasting performance than existing models. |
Å°¿öµå(Keyword) |
½Ã°è¿ ºÐ¼®
´Ùº¯·® µ¥ÀÌÅÍ ºÐ¼®
´ÙÃþ ¼öÆò ¿¹Ãø
½ÉÃþ ÇнÀ
±íÀº Àΰø½Å°æ¸Á
Time Series
Multi-variate Data Analysis
Multi-horizon Forecasting
Deep Learning
Neural Networks
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|