Industrial applications of unification morphology
ANLC '94 Proceedings of the fourth conference on Applied natural language processing
Natural languages and the Chomsky hierarchy
EACL '85 Proceedings of the second conference on European chapter of the Association for Computational Linguistics
COLING '94 Proceedings of the 15th conference on Computational linguistics - Volume 2
COLING '92 Proceedings of the 14th conference on Computational linguistics - Volume 3
| Hi-index | 0.00 |
We describe how a simple parser can be built on the basis of morphology and a morphological analyzer. Our initial conditions have been the techniques and principles of Humor, a reversible, string-based unification tool (Pròszéky 1994). Parsing is performed by the same engine as morphological analysis. It is useful when there is not enough space to add a new engine to an existing morphology-based application (e.g. a spell-checker), but you would like to handle sentence-level information, as well (e.g. a grammar checker). The morphological analyzer breaks up words into several parts, all of which stored in the main lexicon. Each part has a feature structure and the validity of the input word is checked by unifying them. The morphological analyzer returns various information about a word including its categorization. In a sentence, the category of each word (or morpheme) is considered a metaletter, and the sentence itself can be transformed into a meta-word that essentially behaves like a real one. Thus the set of sentences recognized by the parser called HumorESK can form a lexicon of meta-words that are processed much the same way as lexicons of real words (morphology). This means that algorithmic parsing step are substituted by lexicon look-up, which, by definition, is performed following the surface order of string elements. Both the finitizer that transforms formal grammars into finite lexicons and the run-time parser of the proposed model have running implementations.