LEARNING MANAGEMENT SYSTEM AND LEARNING MANAGEMENT METHOD

To provide a learning management system and a learning management method that appropriately determine parameters of a subsystem control method in a complex system that cannot be embodied as a simulator and have large changes to past performance data.SOLUTION: A retail agent 301 that determines, by u...

Full description

Saved in:
Bibliographic Details
Main Authors MATSUMOTO KOSEI, ITO KIYOHITO, FUJI DAIKI, ESAKI KANAKO
Format Patent
LanguageEnglish
Japanese
Published 07.05.2020
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:To provide a learning management system and a learning management method that appropriately determine parameters of a subsystem control method in a complex system that cannot be embodied as a simulator and have large changes to past performance data.SOLUTION: A retail agent 301 that determines, by using data stored in a past performance database 53, parameters of an activity proposal model 511, and further determines, by using foothold activity simulation data, parameters of an activity determination model 522 and an activity value evaluation model 521. This enables, in a complex system that cannot be embodied as a simulator and have large changes to past performance data, appropriate determination of parameters of a subsystem control method.SELECTED DRAWING: Figure 3 【課題】シミュレータとして具現化できず、かつ、過去の実績データに対する変化が大きい複雑系システムにおいて、サブシステムの制御方法のパラメータを適切に決定する。【解決手段】小売エージェント301は、過去実績データベース53に記憶されているデータを用いて、活動提案モデル511のパラメータを決定し、さらに拠点活動模擬データを用いて活動決定モデル522及び活動価値評価モデル521のパラメータを決定する。これにより、シミュレータとして具現化できず、かつ、過去の実績データに対する変化が大きい複雑系システムにおいて、サブシステムの制御方法のパラメータを適切に決定することができる。【選択図】図3
Bibliography:Application Number: JP20180205053