Automatic syllabification for Spanish using lemmatization and derivation to solve the prefix’s prominence issue

•We describe the problem of the syllabification when the word has a prefix.•An algorithm that identifies the prefix’s prominence and applies the right syllabification.•The automatic syllabification is possible using a specific lemmatizer.•The algorithm uses the information provided by a knowledge database about derivation.•We use information about prefixes’ current productivity and word frequency.

The syllabification of Spanish’s words follows a few basic rules, but the syllabification of some words deviates from the general rules according to a number of factors described in this paper. Prefixes are major cause of variations on syllabification. Since, in Spanish, prefixes tend to do not integrate into other syllables when they are prominent, the syllabification of words can vary depending on the prominence of the prefixes. This paper shows that, in many cases, the prominence of a prefix can be inferred by means of some morphological and lexical knowledge. This paper proposes a syllabification algorithm that implements the basic syllabification rules and combines them with morphological and lexical information obtained from three sources: a lemmatizer, a derivation database, and the Corpus de Referencia del Español Actual (CREA) of Royal Spanish Academy. Using this additional information, this paper attempts to provide a solution to the problem of taken into account the prefixes according to its prominence for a correct syllabification.