MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractThis dissertation is composed of three papers that each summarize computational investigations in morphology. The first paper studies the relationship between nominal classification and complexity in language. It subsequently introduces a model for this relationship called the association model of nominal classification. This model claims that nominal classification aids in word storage in gendered languages. These claims are supported by data from German, French, English and Iraqw. The second paper investigates quantitative approaches to measuring the productivity of affixes in Swahili. It subsequently introduces a novel model of measuring productivity called the cumulative root ratio. This model gives a story for the variables that determine whether an affix is productive, and data come from corpus data of Swahili. The third paper studies the relationship between the semantic cohesion of derived words, and the meaning of the affixes which they contain. This study introduces the idea of semantic class coherence and argues that this is correlated with word decomposition in lexical access, and likewise is a prerequisite for affix productivity. This model is supported by data from English word vector space models, along with other corpus data from WordNet and Celex. These three papers each are examples of original research that study human language at the word level using data driven methods. This method of employing computational modeling and machines to investigate human language allows us to better understand the ways in which humans can interact with, acquire, and produce language.
Degree ProgramGraduate College