Discovering linguistic knowledge by converting printed dictionaries of minority languages into machine readable dictionaries

摘要

The goal of the project presented in this paper is to explore the linguistic knowledge hidden in printed dictionaries of minority languages. Firstly, the printed dictionary has to be converted into a machine readable dictionary. The second step is to make use of existing language processing tools to discover the hidden knowledge. To illustrate the proposed idea, a version of an English-Penan dictionary is used as the case-study. It appears that even with a small amount of data, some interesting information, like the first list of functional words, some collocations, and an insight of the morphological structure of the Penan language can be discovered.