Towards the Lemmatisation of Polish Nominal Syntactic Groups Using a Shallow Grammar
ukasz Degórski · 2011
While morphological analysers and taggers usually assign lemmata to wordforms, those tools focus on single words. For some tasks a tool that lemmatises (and thus normalises) whole phrases would be more appropriate. The paper presents, discusses and evaluates a set of tools to lemmatise nominal groups, based on a shallow grammar for Polish. The tools reach an overall success rate of over 58%, and almost 83% on the nominal groups that are correctly recognised by the grammar. The approach should be portable to other languages, especially those morphologically rich.