20.500.12556/DiRROS-31563 Možnosti uporabe umetne inteligence v rodoslovju: metodološki premislek na primeru slovenske emigrantske družine Ložar/Lozar iz Elyja v Minnesoti Možnosti uporabe umetne inteligence v rodoslovju Prispevek predstavlja metodološki premislek o možnostih uporabe umetne inteligence (UI) v rodoslovju na primeru slovenske emigrantske družine Ložar/Lozar iz Prigorice pri Ribnici, ki se ob koncu 19. in na začetku 20. stoletja preseli v Ely v Minnesoti. Izhodišče je Genealoški dokazni standard (GDS), ki določa "razumno izčrpno iskanje" virov, natančno navajanje, analizo in korelacijo informacij, razreševanje nasprotij ter oblikovanje preverljivega sklepa. Najprej pokažemo, kako klasična rodoslovna metoda, zasidrana v zgodovinopisni kritiki virov, vodi do dokaza o sorodstvenem razmerju med Frankom Lozarjem starejšim in Frankom Lozarjem mlajšim. Nato konceptualno obravnavamo tri skupine orodij UI – pogovorne sisteme, OCR/HTR in prepoznavanje imenovanih entitet (NER) – ter na študiji primera družine Ložar/Lozar pokažemo, kako lahko dopolnijo iskanje, transkripcijo in strukturiranje gradiva. Sklepno poudarimo, da morajo digitalni izhodi ostati hipoteze, ki dokazno veljavo pridobijo šele z verificiranjem po načelih GDS, in zagovarjamo hibridni model, v katerem UI deluje kot infrastrukturna podpora klasičnemu rodoslovnemu raziskovanju. This article offers a methodological reflection on the use of artificial intelligence in genealogy, based on the case study of the Slovenian emigrant family Ložar/Lozar from Prigorica near Ribnica, who settled in Ely, Minnesota, at the turn of the twentieth century. Our point of departure is the Genealogical Proof Standard (GPS), which requires reasonably exhaustive research, accurate citation, analysis and correlation of information, resolution of conflicts, and a sound conclusion. We first show how the classical genealogical method, grounded in historical source criticism, leads to a documented proof of the kinship between Frank Lozar Sr and Frank Lozar Jr. We then conceptually examine three groups of AI-supported tools – conversational systems, OCR/HTR, and named entity recognition (NER) – and discuss how they can complement searching, transcription, and the structuring of sources. We conclude by arguing that digital outputs should be treated as hypotheses that acquire evidentiary force only through GPS-guided verification, and we advocate a hybrid model in which AI functions as infrastructural support to traditional genealogical research. umetna inteligenca digitalna humanistika genealogija slovensko izseljenstvo pogovorni sistemi OCR HTR NER Združene države Amerike artificial intelligence digital humanities genealogy Slovenian emigration chatbots OCR HTR NER United States true false true Slovenski jezik Angleški jezik Neznano 2026-08-05 11:21:49 2026-08-05 11:21:49 2026-08-06 04:01:32 0000-00-00 00:00:00 2025 0 0 Nasl. z nasl. zaslona; Opis vira z dne 12. 2. 2026; str. 254-278 št. 2 Letn. 8 2025 0000-00-00 Zaloznikova Objavljeno NiDoloceno 0000-00-00 0000-00-00 2025-01-01 929.5:004.89 2591-0884 10.54356/MA/2025/EETZ1766 268284675 https://www.pokarh-mb.si/storage/app/media/Moderna_arhivistika_2025_2/13_Hazemali_2025.pdf 1 173d01bb-90af-11f1-af5a-001a4af901a5 https://dirros.openscience.si/Dokument.php?lang=slv&id=48934 RAZ_Hazemali_David_2025.pdf RAZ_Hazemali_David_2025.pdf 1 BD7D28B634374423FAC2A34D7FEF6A23 9c19f98acec7bd87cff06285ebbd67dc9a3d232ead478eed936c359b745a37e6 22f204db-90af-11f1-af5a-001a4af901a5 https://dirros.openscience.si/Dokument.php?lang=slv&id=48935 Pokrajinski arhiv Maribor 0 0 0