Supplementary Materials for "Cross-Dataset Evaluation and Statistical Selection of Machine Learning Models for Soil-Driven Crop Recommendation"

This record contains the supplementary materials associated with the manuscript “Cross-Dataset Evaluation and Statistical Selection of Machine Learning Models for Soil-Driven Crop Recommendation.” The supplementary workbook consolidates the exact 8,200-record Sulu-oriented agronomically constrained synthetic dataset analyzed in the study together with seven-model performance results, cross-dataset benchmark results, Bayesian-search information, and crop-range summary tables. The materials are provided to support transparency and reproducibility of the reported machine-learning experiments. The Sulu-oriented dataset contains 41 balanced crop classes and seven soil-related predictors: nitrogen, phosphorus, potassium, soil temperature, pH, electrical conductivity, and soil moisture. The manuscript is currently under consideration at SN Computer Science.