Free/Open Source Shallow-Transfer Based Machine Translation for Spanish and Aragonese

s_an.nt.bz2 for Spanish and Aragonese, respectively. Available from https://github.com/jimregan/ es-an-sentences Table 1: Example first sentences from the Wikipedia articles “Mar”. es Un mar es una masa de agua salada de tamaño inferior al océano an A mar u o mar ye una masa d’augua salada de grandaria inferior a l’ocián A sea is a body of salt water smaller than an ocean Table 2: Example extract from the DBpedia abstract dataset. es John Joseph Nicholson es un actor, productor, guionista y director de cine estadounidense doce veces nominado y tres veces ganador del Premio de la Academia. En activo como actor desde 1958. an Jack Nicholson (nombre artistico de John Joseph Nicholson) ye un actor y director cinematografico estatounitense, naixito o 22 d’abril de 1937 en Nueva York. Gloss es John Joseph Nicholson is an actor, producer, screenwriter and director of American cinema twelve times nominated and three times winner of the Academy Award. Active as an actor since 1958. Gloss an Jack Nicholson (artistic name of John Joseph Nicholson) is an American actor and director, born 22 April 1937 in New York. tions, both for normalisation of non-standard forms, and for cognate induction; and to build a set of equivalent suffixes, which, as well as functioning as transformations, also served as a means of filtering words for equivalence, and for assigning categories: for example, -dá, -dat, daz and -datz all refer to the same feminine noun (-dat, -datz in the standard orthography), which typically have cognates with the suffix -dad, -dades in Spanish: for example, (uniformidat, uniformidatz (“uniformity”, “uniformities”) in Aragonese, uniformidad, uniformidades in Spanish. Cognate induction was then performed by filtering words by suffix, applying the extracted transformations, and comTable 3: SPARQL query used to extract abstracts.