Keep the numerical wording discrepancy open. In-sample group separation cannot be reported as a validated identification success rate.
The 53/58 wording needs reconciliation against PDF and supplementary material. Do not convert explained variance or visible separation into prediction accuracy.
Results report 58 detected elements and 23 with significant type differences. The abstract instead refers to 53 elements. CVA separates groups in the study sample; the article does not report an external held-out authentication accuracy.
3.1
3.1
Results 3.1; Tables 2–3
Checked against the supplied PDF browser display (French Rum) or publisher HTML (Hopfer) on 2026-09-06. Verified means faithful representation of the source, not independent confirmation of the source's claim. Preservation holds remain on Source.
No access / Whiskey Excerpts Database (USWA)
Verified means this note faithfully represents the material reviewed. It does not mean the author's historical or causal claims have been independently established.
Evidence
Results report 58 detected elements and 23 with significant type differences. The abstract instead refers to 53 elements. CVA separates groups in the study sample; the article does not report an external held-out authentication accuracy.
Evidence type: Paraphrase. This is not a verbatim quotation.
Location and context
Source: No access.
Publisher HTML: Results 3.1; Tables 2–3.
The 53/58 wording needs reconciliation against PDF and supplementary material. Do not convert explained variance or visible separation into prediction accuracy.
Annotation
Keep the numerical wording discrepancy open. In-sample group separation cannot be reported as a validated identification success rate.
Literature Note: 
Citation: