Automatic lexical acquisition from Raw Corpora: an application to Russian

Abstract

This paper presents a methodology for the automatic acquisition of lexical and morpho-syntactic informatio from raw corpora. The system uses information about the inclectional morphology declared by rules and is based on the co-occurrence of different forms of the same paradigm in the corpus. A direct application of this methodology gices very poor precision rates due to rule interaction between paradigms. We present a rule analysis algorithm that solves this problem, giving quite better precision rates, although recall decreases damatically. Finally, we investigat some techniques to raise the recall, achieving recall rates around 67% with a precision of 92%.