A Hybrid CFS Filter and RF-RFE Wrapper-Based Feature Extraction for Enhanced Agricultural Crop Yield Prediction Modeling

Abstract

The innovation in science and technical knowledge has prompted an enormous amount of information for the agrarian sector. Machine learning has risen with massive processing techniques to perceive new contingencies in agricultural development. Machine learning is a novel onset for the investigation and determination of unpredictable agrarian issues. Machine learning models actualize the need for scaling the learning model’s performance. Feature selection can impact a machine learning model’s performance by defining a significant feature subset for increasing the performance and identifying the variability. This paper explains a novel hybrid feature extraction procedure, which is an aggregation of the correlation-based filter (CFS) and random forest recursive feature elimination (RFRFE) wrapper framework. The proposed feature extraction approach aims to identify an optimal subclass of features from a collection of climate, soil, and groundwater characteristics for constructing a crop-yield forecasting machine learning model with better performance and accuracy. The model’s precision and effectiveness are estimated (i) with all the features in the dataset, (ii) with essential features obtained using the learning algorithm’s inbuilt ‘feature_importances’ method, and (iii) with the significant features obtained through the proposed hybrid feature extraction technique. The validation of the hybrid CFS and RFRFE feature extraction approach in terms of evaluation metrics, predictive accuracies, and diagnostic plot performance analysis in comparison with random forest, decision tree, and gradient boosting machine learning algorithms are found to be profoundly satisfying.

A Comprehensive Review of Crop Yield Prediction Using Machine Learning Approaches With Special Emphasis on Palm Oil Yield Prediction

“…In CART, information gain, Gini Diversity Index (GDI) and gain ratio are used to split the attributes. RF is a powerful tool for the prediction of yield, which has been applied to agricultural research [46] [42] [53] [56] [24]. It generates a wide range of regression trees that are produced by a large set of decision trees for computing regression [75].…”

“…The performance of a model can be defined by evaluation metrics. Evaluation measure plays a significant role because of their capability in differentiating among the outcomes of different learning models [53]. There are various performance metrics applied to evaluate the performance of the regression model including mean absolute error (MAE), mean squared error (MSE), root mean square error (RMSE), determination coefficient (R-squared) and mean absolute percentage error (MAPE).…”

“…The determination coefficient is used for measuring the accuracy of the fit of the regression framework that reflects how the developed framework is superior to the baseline framework [86]. MAPE is used to calculate the average of the percentage errors that determines how far the prediction of the model deviates from its corresponding outcomes [53]. The machine learning based classification algorithms for crop yield prediction are evaluated by accuracy [87] [88], precision [89], recall [89], sensitivity [90], specificity [90], and F1 Score [91].…”

“…Maya et al[41] compared the performance of SVR, RF, ANN and K-NN and claimed that the RF algorithm is superior by achieving the highest accuracy. Dhivya et al[53] proposed the implementation of a hybrid-based feature selection technique by coupling of correlation type filter CFS along with a wrapper RF-RFE based on recursive feature elimination.…”

Performance Assessment of Certain Machine Learning Models for Predicting the Major Depressive Disorder among IT Professionals during Pandemic times

“…A classical feature selection approach was proposed for blood cell disease recognition [ 35 ], and a robust feature selection approach was proposed for the application based on welding defects detection [ 36 ]. The importance of optimization was illustrated [ 37 ], and a hybrid approach of feature selection was proposed for the application related to agriculture [ 38 ].…”