Architecture development with measurement index for agriculture decision-making system using internet of things and machine learning

Nowadays, the agriculture decision-making system has become essential for field monitoring. In that aspect, the Internet of Things (IoT) and Machine Learning (ML) are the most emerging technologies which can provide precision, intelligence, and agriculture decision-making system to yield better results. These technologies help in increasing agricultural production and enhancing operational efficiency. Due to sudden changes in weather conditions studying various parameters are either affected or influenced, the prediction analysis has been impractical and difficult. In such cases, using few intelligent systems like IoT and ML can provide feasible alternative solutions. In this paper, a novel architecture development is being proposed for agricultural decision-making systems using IoT and ML. The performance metrics of various ML algorithms in the field of agriculture are examined and analyzed in this study. Decision Tree (DT) has shown superior performance over the conventional methods like Support Vector Machine (SVM), and Random Forest (RF) about agriculture sensor data. Simulation results show that the proposed development of architectural measurement index for agriculture decision-making system has a maximum accuracy value of 98%, minimum Mean Absolute Error (MAE) of 0.07%, Mean Square Error (MSE) of 0.06%, R-Squared parameter of 99%, and Root Mean Square Error (RMSE) of 0.002% for detecting crop production in IoT-ML agriculture decision-making system. Also, significant experiments have been carried out to evaluate Measurement Index (MI) with less error rate for agriculture decision-making system.