METHOD FOR PREDICTING ASSOCIATION-RELATED INFORMATION BETWEEN ENTITY-PAIR BY USING MODEL FOR ENCODING TIME SERIES INFORMATION, AND PREDICTION SYSTEM GENERATED BY USING SAME

• Main Authors: KOO, Hee Jung, PARK, Wonpyo, HAN, Seokjin, KIM, Tae Yong
• Format: Patent
• Language: English; French; Korean
• Published: 14.09.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTEDFOR SPECIFIC APPLICATION FIELDS; PHYSICS

The present invention relates to a method for generating a prediction model that is generated by using a value output from a model for encoding time series information as training data and learning the training data to predict association-related information between entities, and a prediction system for performing prediction by using the generated prediction model. La présente invention concerne un procédé de génération d'un modèle de prédiction qui est généré par utilisation d'une valeur délivrée à partir d'un modèle de codage d'informations de série chronologique en tant que données de formation et apprentissage des données de formation pour prédire des informations relatives à une association entre des entités, ainsi qu'un système de prédiction destiné à effectuer une prédiction à l'aide du modèle de prédiction généré. 본 발명은 시계열적 정보를 인코딩하는 모델에서 출력되는 값을 학습 데이터로 사용하고, 개체 사이의 연관성 관련 정보를 예측하도록 학습 데이터를 학습함으로써 생성되는 예측 모델의 생성 방법, 그리고 생성된 예측 모델을 사용하여 예측을 수행하는 예측 시스템에 관한 것이다.