TITLE:
Principles Derived from Neurolinguistics of Brain for Design of Translation Machines
AUTHORS:
Zi-Jian Cai
KEYWORDS:
Language, Three Neurolinguistic Processes, Machine Translation Principle, Semantic Dictionary, Grammar, Statistical Translation
JOURNAL NAME:
Open Access Library Journal,
Vol.3 No.6,
June
8,
2016
ABSTRACT: Recently, it was extended by the author the declarative/procedural model
to a new semantic/ syntactic/episodic model for language, extended to encompass
the sentential meanings to the neural linguistic processes. In this article, it
is applied this new semantic/syntactic/episodic model of brain to derive three
feasible principles to direct machine translation respectively. First, it is
necessary to establish the dictionary for translation of words and phrases. Second,
it is also necessary to read out the grammar of language to be translated from
and to comply with the grammar of language to be translated into, arranging
such parts of speech as noun, verb and adjective into order. Third, it is in
further necessary to determine one correct meaning of some words of multiple
meanings by matching them with statistical associations with others. Whereas,
due to the lack of scientific guidance from neurolinguistics, it has mostly
been adopted two linguistic processes in the present machine translation,
either with only word and grammar translation, or with only word and
statistical translation, and therefore has been unsatisfactory. Through
comparison, it is pointed out that the machine translation with three
principles would exceed the human brain in all three linguistic aspects
respectively. In this regard, herein it is newly formulated the three
principles derived from the semantic/syntactic/episodic neurolinguistic model
of brain for machine translation. Prospectively, it is a significant
progression leading the new technological leap of artificial intelligence with
acquisition of ability in natural language equal to and even superior to that
of human brain.