HLT4LL: John Nerbonne
Bio
John Nerbonne studied Linguistics and Computer Science at the Ohio State University and then worked in industry for eight years (Hewlett Packard Labs and the German Artificial Intelligence Center) before becoming professor of Computational Linguistics and chair of Humanities Computing (alfa-informatica) in Groningen in 1993. He’s worked on a range of theoretical and applied topics in computational linguistics, including grammar development, semantics, natural language interfaces, computer-assisted language learning, language acquisition simulations and detecting syntactic differences in corpora. He has focused most recently on computational tools for analyzing pronunciation differences, where he has contributed a number of techniques and refinements to dialectology. Nerbonne is past president of the ACL (2002), a member of the Royal Netherlands Academy of Arts and Sciences (KNAW), and a Humboldt prize winner. He is currently the president of the European Association for Digital Humanities.
A CALL program for Runyakitara morphology
The talk reports on work in collaboration with Dr. Fridah Katushmererwe (Makerere University, Kampala) and undertaken to support the learning of morphology in Runyakitara, a Bantu language. The major practical objective was to help learners with mother tongue deficiencies to enhance their knowledge of grammar and acquire writing skills in Runyakitara. The system currently focuses on nouns and employs finite state morphology in order to generate a large base of exercise material without extensive tuning by teachers. Language learners used the system over ten sessions, and their improvements were charted. Besides this empirical evaluation, we also sought the opinions of Runyakitara experts about the system (as a judgmental evaluation). Results from the evaluation study indicate that the system has the ability to assess users’ knowledge of Runyitara and to enhance grammar and writing skills in the language. This computational morphology can be utilized to provide more exercises in morphosyntax and the CALL system can support other interested learners of Runyakitara. Finally, we may have identified yet another user group who can benefit from CALL, i.e. emigrant populations with who wish to maintain their language skills in indigenous languages.