Segmentación y análisis morfológico de textos en español utilizando el sistema SMORPH

Zusammenfassung

This paper describes the SMORPH system -a tool for building electronic dictionaries, morphological analysis and tagging of running texts- and how it is currently used to build a Spanish dictionary and to process Spanish texts. The system has shown to be efficient and deals, in a declarative fashion, with several typographical and morphological phenomena (punctuation and typographical rules, prefixes and suffixes, compounds, unknown words, etc.) which, though very usual in running texts,are not taken into account by most grammar building tools and parsers. SMORPH is part of the GRIL works on the study and development of tools for automatic linguistic processing of texts. It must be then considered as a module integrated in other projects (syntactic analysis, etc.).