INTERPOLATE GENERALIZED RANDOM DIVERGENCE COCKROACH AGENTIC AI FOR EFFICIENT CROP RECOMMENDATION

Abstract

Smart agriculture is the fast developing multi-disciplinary field that encompasses the knowledge from agriculture. Crop recommendation is an essential aspect of agriculture for helping the farmers in making suitable decision for attaining the optimal yield and profitability. Many researchers used Machine Learning(ML) and Deep Learning (DL) methods for performing efficient crop recommendation. But, the accuracy level was not improved and time complexity was not reduced by the existing methods. In order to address the above mentioned issues, a new model termed an Interpolate Generalized Cockroach Agentic AI (InGCAg) model is introduced. At first, gather the agricultural data samples. To perform preprocessing, missing information is handled as well as outliers are detected. After that, the relevant features are selected by Rand Divergence Cockroach Optimization process from pre-processed database. Then, Agentic AI Data Classification is used in proposed InGCAg model to recommend and suggest the crops to the particular location with higher accuracy. Finally, the fine tuning is carried out by using stochastic adaptive moment algorithm to reduce the classification error for attaining better crop recommendation results. Experimental analysis of InGCAg model is carried out using various parameters. Outcome of proposed InGCAg model achieves higher crop recommendation accuracy with minimal time consumption than traditional schemes. The result of InGCAg model demonstrates an improvement in accuracy by 98%, precision by 97%, and F1 score by 97%, along with a 97% increase in recall. In addition, it reduces crop recommendation time by 81ms, RMSE by 0.056when compared to traditional deep learning approaches.