From Characters to Meaning: Challenges and Limitations in the Semantic Analysis of Church Slavonic Texts

This chapter explores the challenges of applying computational methods to Old Church Slavonic and Church Slavonic texts, focusing on the impact of non-standard character encoding and Unicode interoperability on semantic analysis. Developed within the ITSERR project, the study examines the creation of tools and workflows for processing texts from the Cyrillomethodiana portal, addressing issues related to Private Use Area (PUA) characters, font dependencies, and digital text standardisation. Through the development of a Unicode mapping system, a text converter, and the DIACU dataset, the research demonstrates how the normalisation of historical Slavic texts can improve their integration into Natural Language Processing (NLP) environments and semantic retrieval systems. The results highlight the importance of standardised digital resources for the preservation, accessibility, and computational analysis of Slavic religious corpora, providing a foundation for future linguistic, philological, and theological investigations.