University Links: Home Page | Site Map
Covenant University Repository

Some Salient Issues in the Unsupervised Learning of Igbo Morphology

Iheanetu, O.U. and Oha, O. (2017) Some Salient Issues in the Unsupervised Learning of Igbo Morphology. In: Proceedings of the World Congress on Engineering and Computer Science 2017 Vol II , October 25 - 27, 2017, San Francisco USA.

[img] PDF
Download (962Kb)


The issue of automatic learning of the morphology of natural language is an important topic in computational linguistics. This owes to the fact that morphology is foundational to the study of linguistics. In addition, the emerging information society demands the application of Information and Communication Technologies (ICT) to languages in ways that demand human-like analysis of language and this depends to a large extent on the ability to undertake computational analysis of morphology. Even though rule-based and supervised learning approaches to the modeling of morphology have been found to be productive, they have also been discovered to be costly, cumbersome and sucseptible to human errors. Contrarily, unsupervised learning methods do not require the expensive human intervention but as in everything statistical, they demand large volumes of linguistic data. This poses a challenge to resource scarce languages such as Igbo. Furthermore, being a highly agglutinative language, Igbo features certain morphological processes that may not be easily accommodated by most of the frequency-driven unsupervised learning models available. this paper takes a critical look at some of the identified challenges of inducing Igbo morphology as a first step in devising methods by which they can be addressed.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Igbo; Igbo morphology; Rule-based learning; Unsupervised learning; Computational morphology
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science
Depositing User: Mrs Patricia Nwokealisi
Date Deposited: 05 Mar 2020 13:05
Last Modified: 05 Mar 2020 13:05

Actions (login required)

View Item View Item