Language modeling

Definitions of theory, model and algorithm.

Theory, Model, Algorithm - Differences and Common Features.

LECTURE 3

Main points:

1. Definitions of theory, model and algorithm.

2. Language modeling.

3. Translation as an object of linguistic modeling.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Theory is the broadest and the most general of the terms. Theory is defined as “...a structure in which each step depends on preceding steps.... The whole structure rests upon observation and theoretical assumption”. (A. Godman, E. M. F. Payne. Longman Dictionary of Scientific Usage. - M. 1989.). A theory is used for description, explanation, and classification of observed events or for the prediction of future events. A theory is always a regularity, it is based on known laws and provided guidance for the discovery or formulation of other (unknown) laws.

What theory is? It:

1. Identifies all observed events (both already known and new).

2. Explains their meaning (formulate the concepts of the events).

3. Establishes interrelation and interdependence of the events.

4. Predicts future events.

Any theory usually begins with the formulation of a hypothesis.

Hypothesis is a scientifically proved assumption either about an event which cannot be directly observed or about a regularity explaining the behavior of a known set of events. (G. Miram. Translation Algorithm. - Kyiv, 1998).

A model is an abstract logical structure which describes an object or a system by analogy and is intended for the explanation of their behavior. The model is a structure outside the object or system being modeled. (Мороховская Э. Я. Основные аспекты общей теории лингвистических моделей. - Киев, 1975).

Language modeling is a method of formulation of certain general hypotheses about the structure of a language as an abstract semiotic system as well as a method of experimental evaluation of the correctness of these hypotheses.

There are two approaches to any modeling (including language and translation modeling) - modeling by deduction and by induction.

Deduction is formulation of certain sequences of abstract schemes being more or less close approximation of real facts and phenomena of a language and evaluation of the correctness of these schemes by application to real language structures.

Models produced by deduction are called synthetic or generative. (G. Miram, p. 27).

Induction is formulation of abstract model schemes on the basis of specific language material. The correctness of obtained model patterns for the language as a whole is also checked by application of the model to real language formulation. Models produced by induction are called analytical. (G. Miram, p. 28)

Paradigmatic models describe the principles of combining the elements into sets in a language, whereas the syntagmatic ones describe the relation between the elements.

An algorithm is a set of instruction or precisely specific operations comprising certain procedure. Speaking about translation which we understand as a process we may say that to show how translation “works” we need:

1. a theory to tell us what the observables and general principle of their interrelation are

(i. e. on what basis they “operate” in such a way that a translation(written or spoken) is produced).

2. a model to show the system of observables, i. e. their interrelation and hierarchy if any;

3. an algorithm to show what steps are to be taken to obtain a translation of a source text into a target one.

At a certain level translation is indeed a well defined problem which can be solved using algorithms.

Дата добавления: 2014-01-11