A Finite-State Morphological Analyzer for Central Alaskan Yup'ik

Abstract

This thesis presents a detailed description of the design and implementation of a finite-state morphological analyzer for Central Alaskan Yup'ik (ISO 639-3: esu). Using a dictionary and a grammatical description of the language, I implemented the grammatical processes with a test-driven development approach using test-by-generation and test-by-analysis. I evaluated the morphological analyzer on two sources of texts, a collection of culture books and the Yup'ik Bible. The evaluation results show the analyzer has a parsing coverage of 89.1% of tokens in the culture books corpus and 90.4% of tokens in the Yup'ik Bible corpus. My hope is that this finite-state morphological analyzer may be integrated into other language technologies and becomes beneficial to the Central Alaskan Yup'ik community and future language learners.