Using frames to disambiguate prepositions

In natural language processing (NLP), disambiguation is the procedure used to solve name conflicts of polysemic concepts (different meanings); in fact, phrase disambiguation is a problem not totally solved in NLP. Several disambiguation types exist; for instance, noun disambiguation (bank is a word with several meanings) or when a word is both used as a noun and as a preposition. Work herein reported describes an algorithm and its implementation to disambiguate prepositions in Spanish phrases; some of them are: a {to} (Translation to English are in bold for explanation purposes; they give only one of the senses of the Spanish word, which usually have multiple meanings.), ante {before}, bajo {under}, cabe {fits}, desde {from}, contra {against}, en {in}, por {by}, según {according to}, sin {without}, entre {between}, so {under}, con {with}, hacia {towards}, sobre {about}, hasta {until}, de {of}, tras {after}, para {for}. This tool will be used in a text analyzer (being built) that converts a text document to its corresponding ontology representation, with the goal of allowing another program to use the information to answer non trivial questions (**).

► Disambiguation of prepositions is performed. ► It processes text (documents) in Spanish. ► Frames are used to help disambiguation and as carriers of semantics.