Distant reading of Soviet diaries

Abstract This article presents a distant reading approach to diary entries collected in the Prozhiito project, a virtual archive comprising over 380,000 diary records in Russian and other languages. The research addresses three key questions: the genre classification of diary texts, the reflection of public discourse in diaries, and language change in emotional diary narratives. We investigated patterns and historical processes using quantitative methods, including topic modeling, collocation analysis, and automatic classification.