Dr. Santos Gracia Villar, president of the Fundación Universitaria Iberoamericana (Ibero-American University Foundation, FUNIBER), together with Dr. Luis Dzul López, rector of the Universidad Internacional Iberoamericana de México (International Iberoamerican University of Mexico, UNINI Mexico), an institution that is part of the Foundation’s university network, are participating in an innovative study that combines data on soil characteristics with machine learning techniques to optimize crop selection and improve agricultural production.
Agriculture is a fundamental pillar of the economy and livelihood of many regions. The right choice of crops is essential to maximize the productivity and profitability of farms. However, farmers face significant challenges in selecting the most appropriate crops for their land, as the wrong choice can reduce yields and lead to food shortages.
Traditionally, crop selection has been based on farmers’ empirical experience and expert advice. However, this approach can be subjective and does not always consider all the environmental and soil variables that influence crop growth.
The study in question introduces an innovative approach that integrates detailed soil information with machine learning models to recommend the most suitable crops. A machine learning model called RFXG was developed, combining Random Forest and Extreme Gradient Boosting algorithms to analyze parameters such as moisture, nitrogen, potassium, phosphorus, pH, precipitation, and soil temperature.
To validate the effectiveness of the RFXG model, relevant data was collected and preprocessed, and its performance was compared with other machine learning models, including classifiers such as Extra Trees, Multilayer Perceptron, Random Forest, Decision Trees, Logistic Regression, and Extreme Gradient Boosting. Hyperparameter optimization techniques and K-fold cross-validation were applied to ensure the robustness of the model.
The results of the study are promising. The RFXG model achieved 98% accuracy in crop recommendation, outperforming the comparative models. This high level of accuracy indicates that the model can provide immediate and accurate recommendations, making it easier for farmers to make informed decisions about which crops to plant based on the specific characteristics of their soil and environmental conditions.
The implementation of this approach has significant implications for agriculture. By using machine learning models that consider multiple soil and environmental variables, farmers can optimize crop selection, increase yields, and improve the sustainability of their agricultural practices. In addition, this method can contribute to food security by reducing the likelihood of food shortages resulting from inappropriate crop choices.
If you would like to learn more about this study, click here.
To read more research, consult the UNEATLANTICO repository.
The Ibero-American University Foundation (FUNIBER) promotes various master’s degree programs in the field of the environment, such as the Master’s Degree in Environmental Engineering and Technology. This program is ideal for those who wish to actively contribute to environmental protection and sustainable development by acquiring practical skills and up-to-date knowledge in the field of environmental management.
Don’t miss this opportunity to advance your professional career and make a difference in the environmental field. Enroll now in this master’s program and become a leader in sustainability.