Release of German Morphological Analyzer SMOR
In collaboration with the universities of Hildesheim and Stuttgart, CIS has released SMOR, a high-coverage morphological analyzer for German.
A morphological analyzer determines the base form and the morphological features of an inflected word. For example, it determines that "ranntest" is the second person singular past tense indicative form of "rennen". Most natural language processing tasks require access to lemma and morphological features for good performance. Thus, morphological analyzers are critical components of natural language processing pipelines.
SMOR was developed at the Institut für Maschinelle Sprachverarbeitung (IMS), University of Stuttgart, in the years 2002 to 2013 and is now jointly maintained by the Universities of Hildesheim (Prof. Ulrich Heid, Institut für Informationswissenschaft und Sprachtechnologie), Munich (Dr. Helmut Schmid, Centrum für Informations- und
Sprachverarbeitung) and Stuttgart (Prof. Jonas Kuhn, IMS).
SMOR is free for noncommercial use.