Processed Row-Level Data for Trustworthiness Auditing of Machine-Learning-Based Fixed-Bed Biomass Pyrolysis Yield Prediction

This restricted record contains processed row-level analytical datasets and record-level model outputs supporting the study “From Accuracy to Trustworthiness: Grouped Validation, Applicability-Domain Mapping, and Reliability-Aware Optimization for Machine-Learning-Based Fixed-Bed Biomass Pyrolysis Yield Prediction.” The deposited files include cleaned and harmonized analytical datasets, target-specific processed datasets, out-of-fold predictions, conformal prediction intervals, applicability-domain scores, record-level eligibility outputs, and detailed scenario-screening results. The original source workbooks were obtained from the supplementary information of Abaei et al., “Machine Learning-Driven Pyrolysis Optimization: Capturing Hidden Feedstock Effects on Biochar and Bio-Oil Yields,” Industrial & Engineering Chemistry Research (2025), DOI: 10.1021/acs.iecr.5c03390. Those original source workbooks are not redistributed in this record. Access to the restricted files may be granted on a case-by-case basis for editorial assessment, peer-review verification, and bona fide research-verification purposes. Redistribution, republication, or public posting of the files is not permitted. Access does not transfer any rights in the underlying third-party source materials. Corresponding author: Mousa Nazari, [e-mail]