An intelligent decision support system for crop yield prediction using hybrid machine learning algorithms.

Area of Science:

• Agricultural Science
• Computer Science
• Data Science

Background:

• Digitization is increasingly vital across various sectors, including agriculture.
• Accurate crop yield estimation is crucial for enhancing agricultural productivity, informing financial markets, and ensuring food security.

Purpose of the Study:

• To predict and enhance the accuracy of crop yield forecasting.
• To evaluate hybrid machine learning (ML) algorithms for agricultural applications.

Main Methods:

• Proposed hybrid ML algorithms incorporating stacked generalization, gradient boosting, random forest, and LASSO regression.
• Utilized aerial-intel datasets from a GitHub data science repository for demonstration.
• Employed cross-validation to compare the performance of individual and hybrid ML algorithms.

Main Results:

• The stacked generalization ensemble method achieved the highest accuracy at 88.89%.
• Random forest regressor and gradient boosted tree regression showed accuracies of 87.71% and 86.98%, respectively.
• Experimental results identified stacked generalization as a highly promising performer for agricultural datasets.

Conclusions:

• The proposed stacked generalization ML algorithm statistically outperforms other methods for crop yield prediction.
• The developed approach demonstrates effectiveness and provides farmers with fast, accurate responses.
• Hybrid ML algorithms show significant potential for improving agricultural decision-making processes.