Prediction of Soil Organic Carbon Content Using Machine Learning Based Fuzzy C-Means Clustering

Soil organic carbon is a fundamental component of soil health, in this paper proposed Principle Component Analysis based Fuzzy C-Means clustering and Partial least squares regression (PCA-FCM-PLSR) for predicting the soil organic carbon component. In this research facing the they offered limited insights into the underlying relationships between input variables and the predicted outcome problem. Apply the preprocessing technique on LUCAS dataset for increase the model accuracy of the model, then using the FCM for randomly selected initial cluster centers and assigns the closest samples to these centers. The PCA method is solely utilized for the clustering process. Finally, the Partial Least Square Regression PLSR is utilized for the effective prediction of soil organic component in carbon, PLSR model can built based on the clusters in calibration set that validation sample belonged to in order to validate this clustering modelling technique. This model archive the better outcomes compare to the other existing models such as Root Mean Square Error (RMSE) of 1.20, R ^{^} 2, of 6.800 Ratio of Performance of Deviation (RPD) of 2.70, and Ratio of Performance to the inter quartile (RPI) of 2.850. The existing models are the k-means Partial least squares regression (k-Means-PLSR), Transferability of Different Covariates (TDC) and the Deep Neural Network (DNN). Modify the sentences in present teens