LATENT LINGUISTIC CODES FOR MORPHEMES USING INDEPENDENT COMPONENT ANALYSIS
Krista Lagus, Mathias Creutz, Sámi Virpioja · 2005
We study properties of morphemes by analyzing their use in a large Finnish text corpus using Independent Component Analysis (ICA). As a result, we obtain emergent linguistic representations for the morphemes. On a coarse level, main syntactic categories are observed. On a more detailed level, the components depict potential thematic roles of the morphemes. An interesting question is whether these discovered lower-dimensional representations could be directly utilized in language processing applications. 1.