Loading...
Loading...
Contributors:
Dataset URI:
10.5281/zenodo.22789333This deposit contains supplementary evidence, derived data and computational materials supporting the manuscript “A reproducible method for auditing artificial intelligence performance claims in tax and customs administration”, prepared for submission to MethodsX. The package contains supplementary methodological documentation, a source register, a claim ledger, aggregate numerical inputs, Python analysis scripts, computational checks and benchmark results, figures, and reproduction instructions. Four purposively selected published reports provide worked examples concerning shipping fraud detection, electronic invoice classification, value-added tax assistance and customs product classification. The method examines the mathematical consistency, interpretation and reporting of published AI performance claims. Its computational checks assess the implemented calculations; they do not constitute independent validation of the original studies, a representative survey of the literature, or evidence of operational revenue effects. AI assistance was used for source reading and extraction, coding assessment, arithmetic review, analysis scripting, drafting and document preparation. No independent human double-coding is claimed. The author is responsible for the deposited content. The archive contains no confidential taxpayer or customs microdata and does not redistribute the cited third-party publications. Consult the included README and supplementary documentation for file descriptions, requirements, reproduction instructions and limitations. This is a supporting research deposit, not an accepted or published MethodsX article.
Keyword(s):
artificial intelligence; customs administration; performance evaluation; reporting audit; reproducibility.
MeSH Term(s):
Grants and Funding:
No Grants or Funding Data Available
Associated Publication(s):
No Associated Publication Data Available
Supplementary Data:
No Supplementary Data Available
Related Dataset(s):
No Related Dataset Data Available
The National Library of Medicine (NLM) provides information about biomedical datasets. Inclusion on an NLM Web page does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: NLM Web Policies (nih.gov)