Enhanced Millet Crop Yield Prediction Using an Optimized Hybrid Fuzzy Logic Model

Accurate crop prediction is crucial for optimizing agricultural operations and ensuring food security. Because existing techniques do not consider the intricate interconnections between many environmental factors, predicting crop growth using these approaches is difficult. Some have proposed using fuzzy logic to solve this issue. Fuzzy models have gained fame in fields such as security-related research due to their interpretability. Though they aren't always accurate, expert-based fuzzy models are simple to understand. Although numerically-based fuzzy models are more precise, they are also more difficult to understand and work with. This study presents a hybrid approach to crop prediction using a Fuzzy Network (FN) and the Fuzzy C-Means clustering method. Building a Mamdani rule-based framework, the model employs an expert system approach and the data-driven FCM methodology. The FN technique combines these two approaches to boost performance. To get even better results with the fuzzy set parameters, we use the Crayfish Optimization Algorithm (COA). The performance of the proposed model may be evaluated by comparing it to the standard FCM technique on different datasets. The findings show that our model outperforms others in crop prediction and is more accurate, transparent, and interpretable.