Akademik Veri Yönetim Sistemi
Mersin Üniversitesi Dil ve Edebiyat Dergisi, cilt.22, sa.1, ss.1-20, 2026 (Hakemli Dergi)
The digital transformation of lexicography has fundamentally shifted the discipline from the production of static, human-readable artefacts to the creation of dynamic, machine-readable databases. This article examines the theoretical and methodological foundations of developing machine-readable multilingual dictionaries (MRDs), with particular emphasis on semi-automatic derivation via pivot languages. Drawing upon the experimental results of a research project we conducted, we analyse the efficacy of utilising the Text Encoding Initiative (TEI) guidelines and ISO-24613 (Lexical Markup Framework) to model complex lexical data. By synthesising historical critiques of database models with contemporary standards for lexical interoperability, we argue that, while fully automatic induction of multilingual lexicons remains fraught with semantic ambiguity, a semi-automatic workflow, grounded in rigorous data modelling and human verification, offers a scalable solution to overcome the resource scarcity inherent in many language pairs.