Feature selection and performance assessment of machine learning algorithms for sunflower oil yield prediction

Abstract

Rapid innovation and liberalized market economy in agriculture require a fast and accurate Sunflower Oil Yield Prediction (SOYP). SOYP is a complex task since it depends on multiple factors. Machine Learning (ML) algorithms and the selection of important features could play a significant role in an accurate oil yield prediction. In this study, we developed ML models to predict oil yield by using two different sets of features. Moreover, we evaluated the most relevant features for accurate SOYP. ML algorithms that were used and compared, were Artificial Neural Network, Support Vector Regression, K-Nearest Neighbour, and Random Forest Regressor (RFR). The dataset consisted of 1250 samples of which 70% were randomly selected and were used to train the model and 30% were used to test the model and assess its performance. The results show that the RFR algorithm achieved the highest accuracy for both feature subsets. These research results indicate that ML has great potential for application as an alternative method for genotypic selection.