Hybrid model in machine learning–robust regression applied for sustainability agriculture and food security

<span lang="EN-US">A dataset containing 1924 observations used in this study to evaluate the effect of 435 different independent variables on one dependent variable. Big data has some issues such as irrelevant and outliers. Therefore, focused analysing comparing impact three variable selection based machine learning techniques, including random forest (RF), support vector machines (SVM), Boosting. Further, M robust regression was applied address outliers using M–bi square, M–Hampel, M–Huber. Random M-Hampel results revealed significant from other methods mean absolute error (MAE) 175.33995, square (MSE) 31.8608, average percentage (MAPE) 9.16091, sum (SSE) 89270.45, R–square 0.829511, adjusted 0.82670. Also, these techniques indicated that 8 criteria were lower than Akaike information criterion (AIC) 47.25915, generalized cross validation (GCV) 47.27169, hannan-quinn (HQ) 47.60351, RICE (47.2845), SCHWARZ 51.7099, sigma (SGMASQ) 46.50605, SHIBATA 47.23489, final prediction (FPE) 47.25929. recommended best models are helpful show minimum efficient for big data.</span>