SYSTEM AND METHOD FOR GROSS DOMESTIC PRODUCT PREDICTION BASED ON DEEP LEARNING MODEL

• Main Authors: AHN HOO YOUNG, LEE HO SUNG, YOU TAE WAN, KIM YOUNG MIN, LEE YEON HEE, YEO DO YEOB
• Format: Patent
• Language: English; Korean
• Published: 23.08.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The present invention relates to a system for predicting gross domestic product based on a deep learning model and a method therefor. The method for predicting gross domestic product based on a deep learning model according to an embodiment of the present invention includes the steps of: (a) selecting a real economic index and obtaining data; (b) processing missing economic indicators through time-series data processing on the obtained data; (c) reconstructing the missing-processed time-series data and performing deep learning model training; (d) predicting gross domestic product using multiple models; and (e) determining and disclosing gross domestic product by combining a predicted result obtained in step (d). 본 발명은 딥러닝 모델 기반 국내 총생산 예측 시스템 및 그 방법에 관한 것이다. 본 발명의 실시예에 따른 딥러닝 모델 기반 국내 총생산 예측 방법은 (a) 실물 경제 지표를 선정하고 데이터를 획득하는 단계와, (b) 획득된 데이터에 대한 시계열 데이터 처리를 통해 경제지표 결측을 처리하는 단계와, (c) 결측이 처리된 시계열 데이터를 재구성하고 딥러닝 모델 학습을 수행하는 단계와, (d) 다중 모델을 활용하여 국내 총생산을 예측하는 단계 및 (e) (d) 단계에서의 예측 결과를 조합하여 국내 총생산을 확정하고 공시하는 단계를 포함한다.