Induction of fine-grained part-of-speech taggers via classifier combination and crosslingual projection

Abstract

This paper presents an original approach to part-of-speech tagging of fine-grained features (such as case, aspect, and adjective person/number) in languages such as English where these properties are generally not morphologically marked. The goals of such rich lexical tagging in English are to provide additional features for word alignment models in bilingual corpora (for statistical machine translation), and to provide an information source for part-of-speech tagger induction in new languages via tag projection across bilingual corpora. First, we present a classifier-combination approach to tagging English bitext with very fine-grained part-of-speech tags necessary for annotating morphologically richer languages such as Czech and French, combining the extracted features of three major English parsers, and achieve fine-grained-tag-level syntactic analysis accuracy higher than any individual parser. Second, we present experimental results for the cross-language projection of part-of-speech taggers in Czech and French via word-aligned bitext, achieving successful fine-grained part-of-speech tagging of these languages without any Czech or French training data of any kind.