The initial practical matters of creation formal grammar of Uzbek language

Размещено на http://www.allbest.ru/

Karshi State University

THE INITIAL PRACTICAL MATTERS OF CREATION FORMAL GRAMMAR OF UZBEK LANGUAGE

Abdisait Norov

The senior lecturer

Abstract

uzbek language formal grammar

Formal grammar of natural language is one of the important factors in “man-machine” formation of communication for intellectual systems. This formal grammar based on linguistic modells in a specific natural language area. This article outlines the initial problems in formulating the formal grammar of the Uzbek language. Lingvo-mathematical models relations and linguistic of the Uzbek language are investigated. There are also comments on some of the solutions as well.

Keywords: artificial intelligence, intelligent system, linguistic model, lingo-mathematical model, virtual expand of alphabet, formal grammar.

The main text

Formulating (modeling) long-term experiences and knowledge which accumulated by the humanity, in formal type and creating different methods of presenting this knowledge in computer have become one of the most current issues which today's scientists have been facing. Numerous scientific research and practical works have been made by scientists over the years in order to solve these problems. As a result, new scientific concepts such as “artificial intelligence ”, “intelligent system” or “expert system” have come into our society at the level of modern science in recent years [1, p. 5].

The issues of artificial intelligence in the sphere of research are various in terms of their essence and complexity; individual, specific methods are used in order to solve them. When we say about the research model and methods of artificial intelligence, natural language processing (NLP) or machine translation is at one of the first places [2, p. 130].

In particular, issues such as understanding the text in natural language, analyzing them, reproducing and generating, also, machine translation, creating information search methods with perfection are considered as the urgent problems in recent years. In addition, issues such as differentiating various symbols, manuscript texts and even oral speech with program are included to the research list of artificial intelligence.

With one word, the ling ware needs for any intelligence machine which was created on the basis of artificial intelligence methods. For doing this, we should formulate models which are suitable to intelligence systems, with step-by-step, with particular one- by-one, in particular:

Firstly, it is necessary to create linguistic software for ling ware;

Secondly, lingo-mathematical model should be created for linguistic software;

Thirdly, linguistic model is necessary for lingo- mathematical model.

According to its essence, any model is the scientific theory which is directed to show internal and external characters of objects or real event which are faced in the nature.

Because of being complicated and multilateral of the events of nature, it is impossible to describe fully its whole properties by single model. For this reason mathematic model of any real object has the formal appearance (without having to go beyond certain research).

For example, the alphabet is the formal model of sounds in certain language in writing form. In another word, alphabet is one of the formal linguistic models which are directed to illustrate the sounds in writing form.

In general, there are enough the concepts of linguistics which are correspond with mathematical concepts and practical issues which can be solved by applying directly to mathematical methods.

It is understood that building mathematical model of the language is to express systematically relationship between the various aspects of this language on the basis of mathematic legality.

Linguistic events which are researched in linguistics, include so much complicated aspects, thus it is worked with complicated mathematical models in the non-existent, dynamic and discrete forms in mathematical modeling of them.

Uzbek linguist A.Pulatov (2011) describes creating criteria of information method of Uzbek language, in which he emphasizes that each word must be clearly and one meaningful [3, p. 366].

The concept of “information method” forwarded by A.Pulatov can regarded as equivalent to the concept of “linguistic model”.

Indeed, linguistic model should have characters such as accuracy and uniqueness in order to base for the lingo-mathematical model.

Let's analyze another example, namely, whole set of symbols (or letters) which are necessary to create text in computer in Uzbek language based on Latin graphics, in order to better understand accuracy and uniqueness. In this case, we can use set concept in the meaning mathematic.

Thus, we take alphabet in Latin language as the set, we mark it such as . The special code is separated in computer for each letter of this set, namely, it is encrypted from “A” to “Z” and from “a” to “z”.

If we write set on the basis of mathematic rules conforming growth order of these codes, then the following expression is appropriate:

sponding to [A; Z] set which contains 26 Latin capital letters and [a; z] set which contains 26 Latin small letters, in the (1) expression, then we can write (1) expression in the following:

Then we try to illustrate the Uzbek language alphabet in Latin letters on the basis of (2) as separate set. For doing this, we use “difference two set” concept in the mathematics.

According to this, we have the followings:

It can be seen from (3) expression, “ W” and “w” letters are divided from alphabet in Latin language and two symbols are added in order to create national alphabet. If we write expressions such as

for simplicity, then (3) expression will be in the following:

It should be noted that, (4) set can be considered the union of two subsets: set of letters and set of symbols. But according to the point of Uzbek language, it 1 2 cannot be considered set as the set of letters in full meaning, because corresponding to “C” and “c” symbols exist in them, but they do not exist as the letter in our alphabet, and they are auxiliary symbols which participate to illustrate “CH” and “ch” letters in computer.

Doing thus, (4) expression (namely _z set) is considered lingo-mathematical model of Uzbek national alphabet.

In order to be clear of our opinions, (in order to clear and understandable of our opinions) let's compare each of two models specific to alphabet, namely * linguistic () and lingo-mathematical () models:

According to the linguistic model: = {A, D, E, F, G, H, I, J, K, L, M, N, O, P, Q, R, S, T, U, V, X, Y, Z, O`, G`, SH, CH, NG, a, b, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r, s, t, u, v, x, y, z, o`, g`, sh, ch, ng, '} - total 29+29+1=59 symbols.

According to the lingo-mathematical model:

= {A, B, C, D, E, F, G, H, I, J, K, L, M, N, O,

P, Q, R, S, T, U, V, X, Y, Z, a, b, c, d, e, f, g, h, i, j, k,

m, n, o, p, q, r, s, t, u, v, x, y, z, `, '} - total 25+25+2=52 symbols.

As we can see that, according to the lingo-mathematical model there are 52 symbols in alphabet, it is less 7 symbols than the linguistic model * in alphabet. We can explain that with the following two reasons:

First of all, there is a statement as “every element of a set must be unique; no two members may be identical" in mathematics. According to that, if we add “s” and “sh” letters to (4), then “s” element will repeat twice, it is contrast to the mathematic rule which is accepted against to the set.

Secondly, it can be creating any words related to pure lexicology of Uzbek language with aid of combination of letters existed in (4) expression. Doing thus, the elements of set (4) are enough to insert optional text in Uzbek national alphabet into computer.

It should be noted that, reflecting scientific opinions on the basis of phonograms specific to national alphabet, taking part in international scientific circles, activating intellectually specific to information society are difficult issue in recent years which the technical progress has been developing. It should be required to be widened again its research sphere in order to add Uzbek language to automated information system which is the central syllable of today's modern development. The methods directed to solve this problem are various, one of them is the method of virtual expand of alphabet.

What does virtual expand of alphabet mean?

It is known necessity of additional alphabet which is used as the auxiliary program to create linguistic software specific to modern equipments for programmers. Additional alphabet may consist of set of letters, symbols and different phonetic marks existed in another language.

Thus, if existed orthographic opportunities of being researched language are not enough for new research works, main alphabet will be filled on account of elements of additional alphabet. This alphabet is only alphabet for scientific purposes; it serves as the base for settling some shortcomings belonging to lexicology, phonetics, morphology and syntax fields of language and creating new models in terms of science.

Then we give attention to necessary aspects and practical importance to virtual expanding the alphabet.

1. The practical importance of virtual alphabet of Uzbek language in the lexicology.

In the recent years we can see that the lexica of Uzbek language have been expanding by volume because of technical progress. For instance, the assimilated words (they are not divided) such as “Daewoo”, “Windows”, “Word” and “PowerPoint” as the proper noun to our lexica have already become publicly terms for recent years. It is required to add “W” and “w” letters, also, if there is necessity, “alien” letters participated in another words like that to our alphabet.

2. The practical importance of virtual alphabet of Uzbek language in phonetics.

The great share of necessity to virtual expands of alphabet fits to its phonetics of Uzbek language.

It is known that concepts such as letter, number, and punctuation marks are unknown for the computer, with one word all of them are recognized as the symbol. The special code for each of these symbols are separated only one, these codes are expressed with decimal or sixteen counting system numbers. For instance, number 51 in the sixteen counting system is separated for writing number 3, number 65 for writing capital letter “A” in Latin language, number 97 for writing small letter “a” in Latin language, number 32 for remove the space and so on.

According to that, each of them should be expressed with unrepeated symbols on the basis of transcription principles in order to be pronounced clearly all phonemes in “clear” manner in computer.

In order to clear of our opinion, we give 3-4 problems with exact examples related to that:

1. it is not expedient to use only one letter “ж” in our alphabet for “журнал” and “жийда” words, but it is expedient to use symbols existed correspondingly in computer such as “ж” and “ж” (if it is in Latin language, it will be “j” and “j”) for this;

2. if we are going to create software which can read with right pronunciation the text or separate syllable, it is difficult to use two letter combination (grapheme) which means one sound;

3. if symbols such as “s” which has one code, use instead of “sh” phoneme which is expressed with two codes, or “c” instead of “ch” phoneme, “g” instead of “g`”phoneme in order to not appear excessive problem in computer-modelling of meaning and pattern relations of words (in particular, such as omograph or omophon), the programming process will be easy (it does not have to be exactly these characters, this is the issue to be solved on the basis of mutual agreement, instead of them it may be received any other symbols which are represented by one code);

4. Vowel “о” is pronounced according to its place in the structure of words, sometimes as “и” (директор - директир), sometimes as “а” (океан - акеан), and sometimes as “у” (электр токи - электр туки) in words assimilated from Russian language. Thus, it is useful to create modification of fonts (or accepting existed transcription symbols) which are suitable for them, for various pronounciations of one letter in order to avoid speech rudeness.

In this case, we consider giving some opinions belonging to vowel and consonant sounds which have been causing many debates among the linguists.

In particular, M.Mirtojiev emphasized that the number of speech sounds in Uzbek language are 48 (the number of unlabeled vowel sounds - 10, the number of labial vowel sounds - 6, the number of sonant - 6, and the number of consonants - 26) and noted each of them with transcription symbols, gave their characters [5, p. 128].

Uzbek linguist H.Jamalkhanov (2009) pauses to issue of expanding alphabet with expense of vowel sounds and pointed following ideas [6, p. 106]:

“There are two opinions by issues about the amount of vowel sounds of current Uzbek literature language in Uzbek linguistics. According to the first opinion, the number of vowel sounds of literature language is 6: и, э, а, у, у, о. (The supporters of this opinion are the followings: E.D. Polivanov, K. Borovkov, V.V. Reshetov, A.G. Gulamov, Sh. Shoabdurakhmonov, F. Kamalov, A. Makhmudov, S. Otamirzaeva, A. Abduazizov, A. Nurmonov, Sh. Rakhmatullaev and others). According to the affirmation of supports of the second opinion (X. Doniyorov, B. Tuychiboev, M. Valiev), there is contrast pairs specific to old Uzbek literature language and current kipchak dialects in current Uzbek literature language vocalism, out of it, vocalism of literature language and alphabet issues should be review and should be added additional letters into alphabet”.

The practical importance of virtual alphabet of Uzbek language in Morphology.

It is known that when the indicator of the third person of possession affixes add to words which the last sound are consonant, any phonetic variant will not create, namely китоб + и = китоби.

Let's analyze this by using the words which нг phoneme participates. For instance, the word “жанги” of the III rd form of word “жанг” is separated syllable such as “жа-нги” according to the criterions of literature language. When it is read by syllable, it will give formal pronunciation which is not exact. Doing thus, it may be separated syllable in formal form such as “жан-ги”! We are going to give opinion that if it should be necessary to pronounce clearly, it should be written as the form “жанг-нги”. In this case, if we adjust нг phoneme with ц symbol (letter) which has one code, then this word will become the form such as “жац-ци” in terms of orthographic and it is suitable for the pattern “CVC-CV” in the separating syllable algorithm.

In this meaning, writing words such as word “кунгил” in the form “кунгнгил” or “куццил”, the word “тонги” in the form “тонгнги” or “тоцци”, the word “инга” in the form “ингнга” or “ицца”, the word “дунгалак” in the form “дунгнгалак”' or “дуццалак” and others, is not rude error for formal grammatics. For this purpose, it may be add “ц” instead of “нг”, if it is in Latin language, for example, it will be “Й” instead of “ng” in contradistinction to “н” and “г”.

The above mentioned opinions are the only points belonging to virtual expand of alphabet, we can say that punctuation mark is not enough only in one place in syntax. For instance, let's take a sample such as “мен бугун уам хизматда, уам иззатда” which is participated conjuction “уам” used in the meaning to emphasize parts of sentence we consider that there necessary to be only one punctuation mark which expresses the intonation, after word “бугун”. Because it is required attentiveness, effectiveness, grammatical knowledgeable from the reader who is unaware of text, in order to pronounce intonation with accuracy in this sentence.

In general, the alphabet which we refer, it serves for the task as the “product” for creating new formal grammar by that and lingo-mathematical modeling which is embodied sounds, punctuation marks, numbers, ligature and diacritic signs which are necessary for speech, therefore, it is considered as the necessary event.

The base essence of “man-machine” communicative system is dependent on machine translation system in the end. There is reproducing and generating issue of natural language on the basis of machine translation. In related to this case, the level of quality of machine translation is marked, from one point, with unerring of syntactic structure of the sentence, by using it in its place.

Affirmation 1: According to the formal model the lexical, phonetic, semantic characters and characters in whole spectrum of the sentence join syntactic structure.

Affirmation 2: Because this formal model is only intended for artificial intelligence systems.

Affirmation 3: The natural intelligence invents the world with the chain of sounds, images, and overall, events which have been happened in reality, in first time, the artificial intelligence is only based on the symbolism which is expressed in the mnemonic codes in that case.

Affirmation 4: The syntactic structure of the sentence is the chain of symbols which are concatenated with known legalities in formal model.

Every specific linguistic model which will be created by applied and structural linguistics will be reformed as the lingo-mathematical model on the basis of such rules by mathematical linguistics.

In conclusion, we can say that the formal grammar of Uzbek language based on virtual alphabet above proposed, will serve as the base for creating “man-machine” intelligent communicative system in the future.

References

1. Павлов С.Н. Системы искусственного интеллекта: учеб, пособие. В 2-х частях. Томск: Эль Контент, 2011. Ч. 1. 176 c.

2. Павлов С.Н. Системы искусственного интеллекта: учеб, пособие. В 2-х частях. Томск: Эль Контент, 2011. Ч. 2. 194 c.

3. Пулатов АД. Компьютер лингвистикаси. Т.: “Akademnashr”, 2011. 520 б.

4. Partee, B.H., Meulen, A., and Wall R.E. (1993). Mathematical Methods in Linguistics. Kluwer Academic Publishers.

5. Миртожиев М.М. Узбек тили фонетикаси. Т.: “Фан”, 2013. 424 б.

6. Жамолхонов X,. Узбек тилининг назарий фонетикаси. Т.: “Фан”, 2009. 224 б.

Размещено на Allbest.ru