SALAMA Dictionary Compiler: A Method for Corpus-Based Dictionary Compilation
Arvi Hurskainen · 2008
This paper describes a new approach to dictionary c ompilation, where a text corpus is converted into dictionary using a set of sophistica ted language processing tools. The corpusbased dictionary compilation has well known advantages and significant advances have been made in it recently. However, various problems in i mplementation have restricted its wider application. The paper describes how a comprehensive language analysis system can be used for automating several labour-intensive phases in d ictionary compilation. SALAMA Dictionary Compiler (SALAMA-DC) is based on a comprehensive language analysis system of Swahili. It produces (a) single-word headwords, (b) multiword headwords, (c) various types of cross-references, and (d) a user-defined s election of use examples in context, including controlled random selection, and selectio n based on frequent contexts. Particularly noteworthy is that the example texts can be transla ted to English and attached to each example. It should be noted that the whole process from raw text to dictionary takes place without manual editing in between. The system was tested with three text corpora.