A computational model of Blackfoot noun and verb morphology

This thesis describes the construction of a computational model of Blackfoot word structure. This model was developed so that it could provide a foundation for Blackfoot language technologies such as spelling and grammar checkers, search suggestion generators, paradigm generators for pedagogical purposes, intelligent dictionaries, automated corpus parsers for linguistic research and more. Many Indigenous languages in Canada have been declining in use. In response, many Indigenous communities and activists have implemented revitalization strategies which vary in effectiveness. One way to help language efforts to be more effective is to ensure that tools for research and revitalization are freely available to community members. In the 21st century this can be achieved in part through technology, particularly with the help of the internet, which offers information freely (in most cases) to those who wish to access it. In this thesis I describe the early developments of a project that will be used to augment currently available digital resources and provide a basis for future technology for the Blackfoot language. I use Finite State Transducer technology to develop a computational model of Blackfoot noun and verb morphology and test the model using a corpus of modern Blackfoot text that was constructed from a curated collection of available texts.

Authors

• Kadlec, Dominik Miroslav
• University of Lethbridge. Faculty of Arts and Science

• Blackfoot
• Blackfoot language
• languages
• linguistics
• computational linguistics
• Indigenous languages
• Algonquian languages
• morphology

Identifier

hdl:10133/6635

OAI identifier oai:identifier

oai:opus.uleth.ca:10133/6635