Assigning an Inflectional Paradigm using the Longest Matching Affix
Krister Lindén · 2008
Abstract. Language software applications encounter new words, e.g., acronyms, technical terminology, loan words, names or compounds of such words. In English new compounds are formed as separate words or using a hyphen, but in other Germanic languages and in Fenno-Ugric languages there is no special indicator between the parts of a compound word. In order to add new words to a lexicon, we need to indicate the inflectional paradigm. In this article, a method for assigning the inflectional paradigm to new compound base forms based on the longest matching affix is evaluated. The method is 99.75 % accurate for Finnish compound base forms. 1