Refurbishing a Morphological Database for German

Refurbishing a Morphological Database for German Petra Steiner Institute for Information Science and Natural Language Processing Hildesheim University steinerp@uni-hildesheim.de Abstract The CELEX database is one of the standard lexical resources for German. It yields a wealth of data especially for phonological and morphological applications. The morphological part comprises deep-structure morphological analyses of German. However, as it was developed in the Nineties, both encoding and spelling are outdated. About one fifth of over 50,000 datasets contain umlauts and signs such as ß. Changes to a modern version cannot be obtained by simple substitution. In this paper, we shortly describe the original content and form of the orthographic and morphological database for German in CELEX. Then we present our work on modernizing the linguistic data. Lemmas and morphological analyses are transferred to a modern standard of encoding by first merging orthographic and morphological information of the lemmas and their entries and then performing a second substitution for the morphs within their morphological analyses. Changes to modern German spelling are performed by substitution rules according to orthographical standards. We show an example of the use of the data for the disambiguation of morphological structures. The discussion describes prospects of future work on this or similar lexicons. The Perl script is publicly available on our website. Keywords: morphology, word structure, deep-structure morphological analyses, CELEX, orthography 1. Introduction and related work While the number of existing digital linguistic data in- creases, sustainability is growing more and more impor- tant within the field of language resources. Whereas some approaches favor the exploitation of multi-authored and distributed knowledge such as Wiktionary (Sagot, 2014; Sylak-Glassman et al., 2015), others’ work also uses older and more established sources (Borin et al., 2009). The current project aims at updating a part of CELEX, which is a database of Dutch, English, and German lexical in- formation (Baayen et al., 1995). Besides information on orthographic, phonological and syntactic features, it also contains ample information on word-formation, especially manually annotated multi-level word structures. Current morphological analyzers for German yield flat structures of morphs or word constituents on different lev- els, e.g. SMOR (Schmid et al., 2004), Gertwol (Haa- palainen and Majorin, 1995), MORPH (Hanrieder, 1991; Hanrieder, 1996), TAGH (Geyken and Hanneforth, 2006) but no hierarchical parses which can provide important in- formation for word sense disambiguation. Only W¨urzner and Hanneforth (2013) present an approach for full mor- phological parsing of German, which is, however, restricted to adjectives. By contrast, the German part of the CELEX database comprises word tree information for a lexicon containing words of all parts of speech and is therefore an important source for deep-structure morphological analyses of Ger- man, which are not available elsewhere. The linguistic in- formation is combined with frequency information based on corpora (Burnage, 1995) which makes it useful for au- tomated morphological analysis of unknown words. However, the German part of the database has some draw- backs which impair its usefulness: About one fifth of the 51,728 lemmas contain letters such as ¨a or ß. …

Content truncated to 500 words