PREDICTION METHOD, PREDICTION DEVICE, AND METHOD OF CREATING PREDICTION MODEL
To accurately predict conditions to reach a specific growth stage of a plant.SOLUTION: Provided is a method executed by a processor (21) of a computer, including: a step (S1) of acquiring aeonome information indicating the concentration of an element collected from an individual of a plant and inclu...
Gespeichert in:
Hauptverfasser: | , , , , |
---|---|
Format: | Patent |
Sprache: | eng ; jpn |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | To accurately predict conditions to reach a specific growth stage of a plant.SOLUTION: Provided is a method executed by a processor (21) of a computer, including: a step (S1) of acquiring aeonome information indicating the concentration of an element collected from an individual of a plant and included in the individual of the plant; and a step (S2) of inputting the acquired aeonome information to a prediction model (24) to predict a growth period which is a period in which the individual reaches a specific growth stage. The prediction model is generated by machine learning based on teacher data (D1) including aeonome information collected as an explanatory variable at one or more time points and information concerning the growth period as an objective variable in association with each other in each of a plurality of individuals of a plant.SELECTED DRAWING: Figure 1
【課題】植物の特定の生育段階に至るまでの条件の予測を精度良く実現する。【解決手段】コンピュータのプロセッサ(21)によって実行される方法であって、植物の個体から収集されて植物の個体に含まれる元素の濃度を示すイオノーム情報を取得する工程(S1)と、予測モデル(24)に、取得したイオノーム情報を入力して、個体が特定の生育段階に至る時期である生育時期を予測する工程(S2)とを含む。予測モデルは、植物の複数の個体の各々において、説明変数として1以上の時点において収集されたイオノーム情報と、目的変数として生育時期に関する情報とを対応付けて含む教師データ(D1)に基づいて、機械学習により生成されている。【選択図】図1 |
---|