Enhanced Convolutional Neural Network for Accurate Crop Recommendation System on Climate Data

Agriculture is crucial for economic growth and development, yet crop productivity is frequently undermined by improper crop selection and ineffective identification of crop types. Traditional systems often focus on isolated factors, such as weather or soil conditions, which leads to less accurate crop suitability predictions. This research addresses these challenges by developing a robust crop recommendation system that integrates multiple factors for improved accuracy. This research aims to develop a robust crop recommendation system by addressing these limitations. We propose a comprehensive approach that includes preprocessing with the Min-Max Normalization algorithm and feature selection using an Enhanced Cuckoo Search Optimization Algorithm (ECSO). The chosen features are classified and Improved Convolutional Neural Network (ICNN) algorithm predicts crops accurately. Our model, combining the CS-ICNN framework, offers enhanced recommendations by considering both soil-specific characteristics and environmental factors. Experimental results demonstrate that the proposed CS-ICNN approach achieves superior accuracy, precision, recall, and reduced execution time compared to existing methodologies.