Classification of Horticultural Planting Seasons Based on Weather and Soil Parameters Using Ensemble Learning

Abstract

Determining optimal planting time for horticultural crops remains a critical challenge due to climate variability and the complexity of environmental interactions. Conventional approaches often rely on empirical judgment, which may not adequately capture dynamic changes in weather and soil conditions. This study proposes a data-driven decision support system that integrates environmental parameters to improve planting time recommendations. The approach utilizes a supervised learning model based on ensemble techniques to classify suitable planting periods using features such as temperature, rainfall, humidity, soil pH, light exposure, and nutrient composition. The dataset is preprocessed through cleaning, encoding, normalization, and label transformation to align with tropical climate characteristics. The developed model is implemented within a web-based platform to assist agricultural practitioners in analyzing field conditions and generating recommendations. Experimental results demonstrate that the model achieves an accuracy of 61.3%, with balanced performance across precision, recall, and F1-score. These findings indicate that the proposed system is capable of capturing complex relationships between environmental variables and planting suitability. The system provides a practical and scalable solution for improving agricultural decision-making through structured data analysis.