2- https://doi.org/10.1007/978-981-16-9991-7_6
Estimation of Wheat Yield Based on Precipitation and Evapotranspiration Using Soft Computing Methods
- Jan 1, 2022
- Abdüsselam Altunkaynak +1 more
Abstract Determination of wheat yield is required in terms of human nutrition, wheat supply planning, water resources management, ecosystem sustainability and economic development. Therefore, estimation of wheat yield is of utmost importance. Thus, in this study, the soft computing methods, i.e., fuzzy, support vector machine (SVM) and multilinear regression (MLR), were developed to predict wheat yield. Precipitation (P) and deficit of crop water requirement (DCWR) data, which were acquired for Konya and Kirikkale between 1992 and 2015, were used as inputs. The prediction performance of employed models was evaluated quantitatively utilizing root-mean-square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE) and coefficient of efficiency (CE). The results revealed that the SVM model performed slightly better than fuzzy model based on RMSE and CE. The SVM and fuzzy models outperformed the MLR model which involves restrictive assumptions such as linearity, normality and homoscedasticity.KeywordsEvapotranspirationFuzzy logicEstimationPrecipitationWheat
- # Mean Absolute Error
- # Estimation Of Wheat Yield
- # Coefficient Of Efficiency
- # Mean Absolute Percentage Error
- # Support Vector Machine
- # Support Vector Machine Model
- # Root-mean-square Error
- # Water Resources Management
- # Ecosystem Sustainability
- # Wheat Yield
