A novel ensemble machine learning and time series approach for oil palm yield prediction using Landsat time series imagery based on NDVI

Accurate oil palm yield prediction is necessary to sustain production for food security and economic return. However, there are limited studies on comprehensive mapping accurate using advanced machine learning algorithms. Using multi-temporal remote sensing data, this paper proposed a new approach predict based the normalized difference vegetation index (NDVI) ensemble algorithm. ReliefF algorithm with linear projection was employed select best combination of spectral indices in discrimination. Oil land cover classified random forest (RF) modified AdaBoost A time-series known as walk-forward validation firstly introduced train model 2016-2019 data one-step performed 2020 RF AdaBoost. Result study revealed that (RMSE = 0.384; MSE 0.148; MAE 0.147) outperformed 0.410; 0.168; 0.176). Our research has demonstrated value detailed subsequent by developing novel utilising satellite imagery, learning, NDVI, which will assist decision-makers managing their practices related palm.