Crop fertility detection plays a crucial role in optimizing agricultural practices and maximizing crop yield. In this study, a combination of neural network (NN) and support vector machine (SVM) algorithms is employed to develop an efficient and accurate system for crop fertility detection. The proposed system utilizes input data derived from various sources, such as soil samples, weather conditions, and crop characteristics. These input variables are preprocessed and fed into the NN and SVM models for training. The NN model employs its ability to learn complex patterns and relationships within the data, while the SVM model utilizes its strong classification capabilities. Through a series of experiments and training iterations, the models are trained to classify soil samples into different fertility levels, such as low, medium, and high. The performance of the models is evaluated using evaluation metrics like accuracy, precision, recall, and F1 score. The results demonstrate that the combined NN and SVM approach achieves high accuracy in crop fertility detection. The neural network captures intricate relationships between the input variables, while the support vector machine effectively separates the data into distinct fertility classes. This combination leverages the strengths of both algorithms and enhances the overall detection accuracy. The developed system holds significant potential for real-world applications in agriculture, enabling farmers to make informed decisions regarding fertilizer application, irrigation, and other farming practices. By accurately identifying crop fertility levels, farmers can optimize resource allocation, minimize environmental impact, and maximize crop yield, ultimately contributing to sustainable and efficient agricultural practices.

Crop classification, classification accuracy, specificity, SVM, NN