Method for learning prediction model for regression prediction of time series data and method for predicting using prediction model

A learning method of a prediction model for a time series data regression prediction and a prediction method using the prediction model according to a preferred embodiment of the present invention enable a prediction model, which is a stacking model for regression prediction of time series data, to...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: CHOO GA YEONG, JEON SO HYUN, PARK SANG JONG, SUNG HYUN JOONG, LEE WON SEOK, KIM DUK HYUNG, LEE MYUNG HO, KOO YOON JEONG
Format: Patent
Sprache:eng ; kor
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:A learning method of a prediction model for a time series data regression prediction and a prediction method using the prediction model according to a preferred embodiment of the present invention enable a prediction model, which is a stacking model for regression prediction of time series data, to be learned, predict a quality using a learned and constructed prediction model, and enable to prevent a deterioration phenomenon of a prediction performance of the prediction model without interfering with a real-time prediction by automatically updating the prediction model. The learning method comprises: a step of acquiring first learning data; a step of learning submodel; a step of acquiring second learning data; a step of learning one meta model; and a step of acquiring one prediction model. 본 발명의 바람직한 실시예에 따른 시계열 데이터 회귀 예측을 위한 예측 모델의 학습 방법 및 예측 모델을 이용한 예측 방법은, 시계열 데이터의 회귀 예측을 위한 스태킹 모델(stacking model)인 예측 모델을 학습하고, 학습되어 구축된 예측 모델을 이용하여 품질을 예측하며 예측 모델을 자동으로 업데이트함으로써, 실시간 예측에 방해가 되지 않으면서도 예측 모델의 예측 성능이 저하되는 현상을 방지할 수 있다.