Notizen

AMBIGUOUS-FIXED 2026-05-15: original frontmatter title 'Profile-Based Re-Identification of Smart Meter Time Series' was a paraphrased framing of the underlying finding — Crossref-resolve of DOI 10.1038/s41598-024-78285-7 confirms the published title is 'The privacy cost of fine-grained electrical consumption data' (same authors, same DOI). Frontmatter volume 'Vol. 14' corrected to Vol. 15 (Crossref publishes 2025-05-19, Scientific Reports Vol. 15). 6-author list filled out (was 'Voyez, A., et al.'). F-121 audit Deep #22. Original DACH-relevant K3-Privacy-Caveat preserved (Smart-Meter-Rollout Wiener Netze). BODY-VERIFIED 2026-05-16 (Deep-27, Editor≠Verifier pass for WFK-1.2.1a #229): Quantitative claims re-checked verbatim against the published article body via PMC open-access full text (PMC12089263). Verbatim quotes from the article: "knowing 5 consecutive electric measures allows to re-identify on average more than 90% of households in our 2.5M half-hourly electric time series dataset"; "rounding to 2 orders of magnitude (i.e., to the closest 100 W) results in 40%+ uniqueness for k=7". Both figures cited in WFK-1.2.1a body ("90 % Re-Identifikationsrate aus 5 Messungen") are CORRECT as stated. Dataset note: 2.5 M households, half-hourly measurements, Enedis France nationwide deployment.