<![CDATA[Agricultural productivity is increasingly challenged by environmental uncertainty caused by fluctuations in temperature, rainfall, humidity, and soil conditions, making accurate crop yield prediction difficult when relying solely on conventional statistical approaches. This study aims to develop a hybrid artificial intelligence framework capable of improving crop yield prediction performance under uncertain environmental conditions by integrating machine learning techniques with fuzzy modelling to better capture nonlinear relationships and ambiguous environmental patterns. The proposed framework employs a hybrid architecture combining a neural learning mechanism with fuzzy inference to process multiple environmental variables, including temperature, precipitation, soil moisture, humidity, and fertilizer usage, as predictive inputs for crop yield estimation. Experimental evaluation was conducted using agricultural datasets containing historical environmental and production records, while model performance was assessed using Root Mean Square Error, Mean Absolute Error, Mean Absolute Percentage Error, and the coefficient of determination. The results demonstrate that the hybrid framework achieved superior predictive accuracy and robustness compared with conventional machine learning models, particularly when handling noisy and uncertain environmental data. Furthermore, the incorporation of fuzzy reasoning enhanced the model's capability to represent uncertainty and improve decision-making reliability in agricultural management. The study concludes that hybrid artificial intelligence approaches provide an effective and scalable solution for supporting sustainable agriculture through more accurate and adaptive crop yield prediction under environmental uncertainty.]]>
Copyrights © 2026
