On Alignment of Textual Elements in a Parallel Diachronic Corpus

Abstract

This paper presents the description of some tools and resources for aligning diachronic parallel texts. We try to emphasize the idea of improving the lexicographical similarity between words and expressions used in old and modern texts, using a special technique of natural language processing namely the BLEU score. The overall result of the research is a package of language tools and resources, which will serve to the automatic alignment of the old Romanian text with the modern text.

Keywords

• text alignment
• parallel diachronic corpus
• word alignment tools
• bleu score