Abstract: Recently, it was proposed by Cai a new semantic/syntactic/episodic neural model of language, newly encompassing the sentential meanings to the linguistic processes, while deriving three feasible principles from this model to direct machine translation respectively. First, it was necessary to establish the dictionary for translation of words/phrases. Second, it was necessary to read out and translate the grammar of language. Third, it was necessary to determine one correct meaning of some words of multiple meanings by matching them with the episodic/statistical associations with others. Later, it was even shown possible to add some new types of episodic associations for application of this linguistic model to machine translation, such as: (1) Classifying the living/natural words and phrases by behavior, adopting both the zoological/ organizational/physical and affective/behavioral/logic/characteristic/changing characters to classify the nouns and verbs, helpful to discerning the episodic associations with other words within the sentence or clause. (2) Classifying the sentence/paragraph into the category of natural/social subjects like physics, biology, art, economy, and so on. (3) Collecting the frequent episodic word-pairs for machine translation. Meanwhile, it was recommended to use the episodic symbolization to mark these characters of words in computer following the corresponding words, such as ⊥,<>, and so on. It is a perspective significant breakthrough from neurolinguistics to machine translation.
Keywords: Language, Semantic/syntactic/episodic neurolinguistic model, Semantic dictionary, Grammar, Episodic association, Episodic symbolization of word, Machine translation.
Title: A Perspective Significant Breakthrough for Machine Translation from Semantic/Syntactic/Episodic Neurolinguistics
Author: Zi-Jian Cai
International Journal of Computer Science and Information Technology Research
ISSN 2348-1196 (print), ISSN 2348-120X (online)
Research Publish Journals
