Modeling the Letters of Nobel Laureate Ivo Andrić as a Knowledge Graph
Year: 2026
Type: dataset
Abstract: This paper presents a workflow for transforming a selected subset of Nobel laureate Ivo Andrić’s letters into a Knowledge Graph (KG). Building on previous work that combined Optical Character Recognition (OCR), Named Entity Recognition (NER), and SPARQL-based contextualization of Andrić’s letters, the present study extends this methodology by introducing an explicit RDF representation of letters. A small corpus of 15 letters written by Ivo Andrić to Zdenka Marković between 1925 and 1939 was first processed using Transkribus. The PAGE XML output was then converted into structured records representing individual letters. These records were enriched with automatically extracted named entities, normalized through post-processing rules, and linked to Wikidata. Finally, the structured and enriched data were transformed into an RDF graph. The resulting KG contains 1,667 RDF triples representing letters, persons, places, organizations, dates, and links to Wikidata. The final output is a queryable RDF graph that supports the structured exploration, analysis, and visualization of Andrić’s letters, providing a more nuanced view of his personal experiences and social networks. This offers a basis for exploring how these factors may have informed Andrić’s literary work. The developed KG can serve as a reusable resource for further research in Digital Humanities.
DOI: 10.48366/r1928003
Language: English
Subfield: Literature and Literary Theory
Field: Arts and Humanities
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7204852025