Publication at Faculty of Mathematics and Physics |
2011
Abstract
In some languages, some words may be written in several ways. Sometimes the variants are equivalent, sometimes not.
There can be standard, nonstandard, dialectical or otherwise marked variants. During automatic processing we need to recognize them all, but at the same time we need a means how to distinguish them, because during synthesis, it is important to select the right variant.
One of the solutions is introduction of so called multiple lemma. We present its possible usa for concrete applications, especially in the field of corpus linguistics.