We present a method for creating a special monolingual dictionary. It is a corpus-driven frequency dictionary whose entries are "verb phrase constructions" (VPCs), not words. It does not contain explicit definitions but corpus examples. A significant part of the process is a"automatic": we use not only natural language processing tools to analyse a text but also specific algorithms which take over some of the typical tasks of the lexicographer, that is, we create the raw entries fully automatically. As a result, the manual lexicographic work is reduced, and the dictionary can be created with a low budget. We are creating a Hungarian dictionary but the core methodology is "language independent". The methodology can be applied for other languages as well. This kind of dictionary can be useful both in language teaching and natural language processing.
© 2001-2025 Fundación Dialnet · Todos los derechos reservados