Software para traducir textos excluyentes y sexistas a incluyentes y no sexistas a través de autómatas

Abstract

This work describes the development of a software whose task is to translate texts written in an exclusive and sexist language to a text in an inclusive and non-sexist language. This software has three processes: first to extract the text from different digital formats (pdf, docx or html), second to translate the text through an automaton that implements the "manual for the non-sexist use of language" and third to return the translated text in a digital document of the same format as the source. The development was carried out through the incremental methodology, from which three increments were obtained: the first to extract the text from the digital document, the second to implement the automaton and translate the extracted text, and the third increment to return the translated text in a format equal to the source document. The implemented automaton was developed in the JFlex tool. The tests carried out are Alpha and consist of translating simple sentences to validate the correct implementation of the rules described in the "manual for the non-sexist use of language". The objective is to help in the learning of this new language that aims to include and recognize women in the newsrooms.