ＬＳＴＭを用いた宮崎マンゴー収穫量の予測（良い予測のための気候因子の考察）

抄録

Mango production is significantly influenced by various climate factors. By understanding the climate factors, accurate forecasting of mango crop yield in Miyazaki will give the benefit for farmers and stakeholders. This study explores the use of Long Short-Term Memory models for predicting the crop yields by examining the climatic factors. The models are built from 7 independent climate factors and cumulative temperature. RMSEs are evaluated as an index of forecasting accuracy. The impact of cumulative temperature on the crop yield has been evaluated, which will be shown in the presentation.