Hunspell for Zazaki: A morphological analyzer and spell checker based on the Vate Journal Corpus

This study aims to develop a rule-based morphological analyzer and spell checker for Zazaki, a low-resource language. The main objective of the study is to model the complex morphological structure of Zazaki and create Zazaki Hunspell, a computational linguistics tool based on the Hunspell architecture, which will support the correct and effective use of the language in digital environments. In this context, the Zazaki morphology (inflectional and morphophonetic) was examined using the Korpusê Kovara Vateyî/ The Vate Magazine Corpus, and a root dictionary (.dic file) and a affix rule file (.aff file) reflecting the affix structure of Zazaki were prepared using the Vate Working Group Dictionaries. The performance of the developed system was evaluated using qualitative methods, primarily quantitative in terms of both morphological analysis (root finding and morphological segmentation) and spell checking (error detection and correction suggestions). The results obtained show that the Hunspell-based rule-based approach can be effective in modeling the basic structures of a morphologically rich language such as Zazaki and in detecting and correcting spelling errors with a certain degree of accuracy. This work is the first comprehensive morphological analyzer and spell checker developed for Zazaki. The possibilities offered by Natural Language Processing (NLP) technologies strengthen the presence of languages in the digital world and make significant contributions to efforts to preserve and revitalize low-resource languages in particular. In this context, the development of Zazaki Hunspell represents not only a technical achievement but also has the potential to increase the digital visibility of Zazaki and support efforts to pass the language on to future generations. The results of this work lay the groundwork for future computational linguistics research on Zazaki and offer an application that could serve as a model for other low-resource languages with similar characteristics. Keywords Zazaki, Kirmanjki, Hunspell, morphological analysis, spell checking