<![CDATA[Determining the appropriate fertilizer type is essential for supporting rice plant growth and optimizing agricultural productivity. However, conventional fertilization practices still rely heavily on empirical judgment and often neglect dynamic soil and plant growth characteristics. This study aims to analyze and compare the performance of several machine learning classification models for fertilizer type determination in rice cultivation. The study employed a computational experimental approach adapted from the CRISP-DM framework using a dataset of 480 records consisting of soil and rice growth parameters, including Nitrogen (N), Phosphorus (P), Potassium (K), soil pH, moisture, and plant height. Five classification algorithms were evaluated, namely Naïve Bayes, K-Nearest Neighbor (KNN), Decision Tree, Support Vector Machine (SVM), and Random Forest. Model performance was assessed using accuracy, precision, recall, and F1-score, combined with Stratified k-Fold Cross Validation. The results showed that Random Forest achieved the best performance with an accuracy of 95.83%, precision of 95.54%, recall of 95.12%, and F1-score of 95.33%. These findings indicate that ensemble learning methods are more effective in handling heterogeneous and multivariable agricultural data than conventional classification approaches. This study contributes to the development of machine learning-based classification analysis for more accurate and data-driven fertilizer determination in rice cultivation.]]>
