On the Meaning preservation capacities in Machine Translation

نویسنده

  • Michael Carl
چکیده

The paper investigates how meaning preservation can be achieved in MT systems. A distinction is made between ontological MT systems and epistemological MT systems. Whereas ontological MT systems perform meaning preservation in the translation process by means of a set of rules that is provided by a system-external world, epistemological MT systems perform an 'understanding' of meaning that is induced through a learning corpus. According to Dum75], theories of meaning can be rich or they can be austere and they can be holistic or molecular. Some approaches to epistemological MT are discussed and classiied according to the terminology introduced. The paper states that "all-purpose" MT is, however, unfeasible. In the MT literature it has often been stated that a translation is valid if and only if the source language text and the target language text have the same meaning. Although this appears intuitively correct, it is unclear what meaning actually is: many people consider meaning as a mental phenomenon. For in-numberable researchers in the eld of computational linguistics and artiicial intelligence meaning is merely a (bracketed) expression that is part of a (meta) language. Some philosophers conceptualize meaning as a compound structure from which parts are in their minds and other parts are in a (physical) world 1. Others see it as an abstract entity that is shared by the members of a (linguistic) society 2 and still others deene the meaning as being dependent on the goals and intentions of the cognitive agent: meaning thus becomes a concept which is relative to an agent's theory (about the world) Sch94]. I do not want to discuss the question of whether the meaning (or parts of it) are in the mind or in the world, rather I will investigate how the meaning preservation requirement is handled in MT systems. I will use the following notations: If a source text source and a target text target are 1 Putnam Put75] deenes meaning as a four dimensional vector containing the 1. syntactic marker 2. semantic marker 3. stereotypes, and 4. the extension of the word. 2 Frege Fre62] diierentiates imagination (Vorstellung), meaning (Sinn) and reference (Bedeutung). Whereas the imagination of a word or sentence is subjective, its meaning is a social phenomenon and the reference is in the world. translations of each other I write T source $ T target. If a source text source and a target text target have an equivalent …

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Towards a Model of Competence for Corpus-Based Machine Translation

A translation is a conversion from a source language into a target language preserving the meaning. A huge number of techniques and computational approaches have been experimented in order to translate natural languages automatically, yet no satisfactory solution has been found. This paper examines approaches to corpus-based machine translation (CBMT). In CBMT, a set of reference example transl...

متن کامل

Meaning preservation in Machine Translation

The paper investigates how meaning preservation can be achieved in MT systems A distinction is made between ontological MT systems and epistemological MT systems Whereas ontological MT systems per form meaning preservation in the translation process by means of a set of rules that is provided by the system external world epistemological MT systems perform an understanding of meaning that is ind...

متن کامل

The Circle of Meaning: from Translation to Paraphrasing and Back

Title of dissertation: THE CIRCLE OF MEANING: FROM TRANSLATION TO PARAPHRASING AND BACK Nitin Madnani, Doctor of Philosophy, 2010 Dissertation directed by: Professor Bonnie Dorr Department of Computer Science The preservation of meaning between inputs and outputs is perhaps the most ambitious and, often, the most elusive goal of systems that attempt to process natural language. Nowhere is this ...

متن کامل

One Step Closer to Automatic Evaluation of Text Simplification Systems

This study explores the possibility of replacing the costly and time-consuming human evaluation of the grammaticality and meaning preservation of the output of text simplification (TS) systems with some automatic measures. The focus is on six widely used machine translation (MT) evaluation metrics and their correlation with human judgements of grammaticality and meaning preservation in text sni...

متن کامل

A Comparative Study of English-Persian Translation of Neural Google Translation

Many studies abroad have focused on neural machine translation and almost all concluded that this method was much closer to humanistic translation than machine translation. Therefore, this paper aimed at investigating whether neural machine translation was more acceptable in English-Persian translation in comparison with machine translation. Hence, two types of text were chosen to be translated...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1998