Randomized trial evaluates a data-driven crop recommendation system in farming, suggesting improved decision-making support.
Key Points
The aim is to develop a machine learning-based crop recommendation system to enhance agricultural productivity by aiding farmers in crop selection based on various environmental factors.
Designed a machine-learning-powered crop recommendation system utilizing supervised learning algorithms including Random Forest and Support Vector Machines.
Conducted detailed analysis on soil chemistry and climatic parameters, employing data preprocessing, feature engineering, and hyperparameter tuning.
Utilized real-world datasets and performed comparative evaluations of model accuracy using metrics such as precision and recall.
Achieved maximum accuracy of 98.7% with the Random Forest classifier in predicting optimal crops.
Evaluated models based on various metrics, including F1-scores, ROC curves, RMSE, and MAE for comprehensive performance benchmarking.
Identified ethical considerations and challenges in scalability, providing a direction for future advancements involving IoT sensors and AI.