Dataset and Scripts for 'Trade-off between Cost and Performance of Earth Observation Data in Olive Tree Health Assessment Using Machine Learning

This dataset supports the research article : Bouslihim, Y., Bouasria, A., Rochdi, A., Haissen, E. B. E., Loczy, D., Orban, Z., & Salem, A. (2025). Trade-off between cost and performance of earth observation data in olive trees health assessment: Digital crop mapping approach using machine learning algorithms. International Journal of Applied Earth Observation and Geoinformation, 142, 104732. doi.org.  Data was collected and processed under the "Olive Trees Health and Yield Prediction through EO Data and Machine Learning" project, funded by the European Space Agency (ESA) within the framework of the EO AFRICA R&D Facility. The dataset includes both raw and processed Earth Observation (EO) data as well as the full set of R scripts used in data preprocessing, machine learning model training, and evaluation. The study investigates how different EO data sources (e.g., Sentinel-2, Mohammed VI satellite, UAV imagery) compare in terms of cost-effectiveness and predictive performance for assessing olive tree health in a Mediterranean agroecosystem. All materials are archived in a single compressed .rar file labeled "SCRIPTS & DATA", which contains: Preprocessed EO datasets used in the analysis Field-based ground truth data for validation R scripts implementing feature extraction, model training, and evaluation pipelines This repository is intended to facilitate transparency, reproducibility, and reuse of the research workflow and data by the scientific community.