Abstract
In this paper we present the Word Manager project which follows a non-standard approach to knowledge acquisition in the domain of natural language processing. The basic idea is that the representation of morphological knowledge should be considered as a knowledge management issue and not only as a parsing problem. As a consequence, Word Manager implements a client/server system which may be characterized as a dedicated object database management system. This approach yields a number of advantages, one of which is that the system lends itself to a ‘bootstrapping approach’ to knowledge acquisition: knowledge is entered in two phases and by two different types of end user. A linguist will first specify rules, and a lexicographer will subsequently make entries. This way, the rules defined by the linguist can be employed for the lexicographer's task: more specifically, the rules can be used to generate entry dialogs, and to create new entries by deriving them semi-automatically from existing entries.
Get full access to this article
View all access options for this article.
