Comparison of Phenological Weather indices based Statistical and Machine Learning Models for Soybean Yield Forecasting in Pantnagar Uttarakhand

Proceeding Paper of ICMMSA 2023 Conference

Comparison of Phenological Weather indices based Statistical and Machine Learning Models for Soybean Yield Forecasting in Pantnagar, Uttarakhand

Authors- Yunish Khan, Vinod Kumar

Abstract- – Early information exchange regarding predicted crop production could play a role in lowering the danger of food insecurity. Predicting crop yields is one of the more difficult tasks in the farming industry. Several investigations have been conducted in the agricultural field to predict increased crop production using the machine learning algorithm Artificial Neural Network (ANN) and statistical model Stepwise Multiple Linear Regression (SMLR). In this study eight multivariate weather-based models including stepwise multiple linear regression (SMLR), principal component analysis (PCA), artificial neural network (ANN) and combinations of them using weather indices and direct weather variables were investigated by fixing 80% of the data for calibration and the remaining dataset for validation to predict soybean yield for Pantnagar, Uttarakhand. Based on the value of R2 (0.95) and nRMSE (7.16%) during calibration stage, the PCA-ANN-W model performed excellent, becoming the best model for soybean prediction compared to other models in the study region. The overall ranking based on the performances of the models can be given as: PCA-ANN-W > ANN-WI > SMLR-W > SMLR-WI ≈ PCA-SMLR-WI > ANN-W > PCA-ANN-WI > PCA-SMLR-W. The study results indicated that PCA-ANN-W and ANN-WI model performed well for the study region as compared to other models.

DOI: 10.61463/ijset.icmmsa-2023.101