Abstract
The article presents the results of a corpus-based study of named entities denoting geographic locations (LOC) in Russian short stories of the 1930s. The research material consists of 100 texts from the Corpus of Russian Short Stories of the 20th century, selected to ensure an even distribution across years (ten stories for each year from 1931 to 1940) and authored by different writers, thus forming a representative sample. Automatic extraction of locations was performed using the Natasha library, normalization was carried out with the large language model Gemma2-9b-it, the results were verified manually. The analysis identified more than 400 unique geographic names; precise coordinates were determined for 331 of them, enabling their visualization on a map. The spatial distribution covers both the USSR and foreign countries, with the highest concentration of locations in the European part of the USSR, Central Asia, and the south of Siberia. A contextual analysis of references to Western countries, often serving as a cultural contrast to the Soviet Union, to Eastern toponyms, appearing sporadically and without pronounced ideological connotations, as well as to the Union republics, depicted as an organic part of the unified Soviet space, was conducted. Fictional locations were virtually absent, reflecting the realist paradigm of literature of the period. The findings contribute to the study of the distinctive features of the past century Russian short prose and may be applied in both literary scholarship and interdisciplinary research, including cultural geography and digital humanities.