CC BY
0ARTIFICIAL SEMANTIC MEMORY OF NOUNS IN UZBEK
LANGUAGE
Zarmasov Sh.
Researcher of Andijan State University https://doi.org/10.5281/zenodo.13682329
Abstract. This article discusses the semantic field of the part of speech of a noun, the connections of the semantic field, the types of paradigmatic relations, and the similar and contradictory features of the members of a paradigm. The relationship of paradigms in the semantic field of nouns is stored in the consciousness of language owners as an entity that is not subject to direct observation. In the artificial semantic memory of computer linguistics, the model of such unobservables is summarized as an entity, universality. However, the process of modeling is carried out by identifying, comparing, sorting and generalizing the "directly observed" feature, the most important peculiarities of the event. In the article, as a linguistic base for the artificial semantic memory of the national corpus of Uzbek language, paradigms in the semantic field of nouns are modeled nationally in a form not specified during direct observation, and recommendations for modeling are given. At the end of a small study, a conclusion is made on what the linguo-cultural world of Uzbek speakers is, such a world should be artificially reflected in the national corpus of Uzbek language.
Keywords: modeling, national model, semantic field, semantic network, artificial semantic memory, lexical meaning groups, paradigm, set-classification, paradigmatic relation, hierarchical relation, nestedness, national corpus of Uzbek language, computer linguistics.
Globalization has accelerated the automation of social life. Today's society strives to receive information in an automated form. A number of studies are being conducted in science to create artificial intelligence systems. In particular, human-computer interaction has been optimized, an electronic corpus of language units, electronic dictionaries and electronic resources of the semantic field have been created.
It is known that the approach to language as a system, the identification of such dialectical categories as generality-specificity in units of language and their interaction leads to the study of the semantic field of the world, which presents a set of infinite paradigms, as a huge systematic area. F. Berezin compares language to a huge fabric, each fiber of which is tightly connected with other fibers. In our opinion, this "huge fabric" can be compared to a semantic field.
The term semantic field was first used by J. Trier. By the term "semantic field" he means "a field or range of concepts". After that, V. Portzig expresses an opinion about "the essential connection of meanings with each other", G. Ipsen - about "the ability of words to form certain groups", A. Jolles - about "the semantic combination of lexical units", F. Dornseif and V. Wertburg - about "the semantic grouping of lexemes". In Uzbek linguistics, the opinions of such famous linguists as A. Nurmanov, H. Nematov, R. Rasulov, E. Begimov, T. Mirzakulov, Sh. Iskandarova, O. Bozorov, B. Kilichev, R. Safarova, H. Khodjieva, M. Narzieva, I. Kochkortoev, and A. Sobirov on the concept of semantic field are of great importance. But the boundaries between "emotional" and "conceptual" meanings, between "meaning" and "definition", between "content" and "reference", between "denotation" and "connotation", between "sign" and "symbol" have not yet been fully defined.
In this article, the issues of identifying the above-mentioned boundaries that constitute the essence of the semantic field based on the dialectic of universality-specificity, the unification of lexemes in the consciousness of native speakers of a national language based on certain patterns, the organization of an artificial semantic field (memory) in computer linguistics based on unification are analyzed on the example of the semantic field of a part of speech of a noun.
Reflection of the world in human consciousness, and then manifestation in language as a secondary image is a complex process, in the subsequent stage of this process, moving away from heterogeneous signs of the object through observation of the feature that affects our senses - "given to direct observation", similar, identical signs are determined, they are compared and generalized as an essence, thus, a concept is formed, and it, in turn, finds its expression in a specific lexeme. Modeling the semantic field in this order allows native speakers to understand the full set of lexical units used in a certain area, use the most necessary of them (those that are needed) in the process of communication, provides a semantic connection of lexical units.
Objects that resemble each other in the semantic field of a noun form single, united groups based on a paradigmatic relationship. Such lexical semantic groups have a form in the consciousness of native speakers that is "not given to direct observation".
Similar and slightly different lexemes are combined into paradigms inherent to the noun, not given to direct observation, different signs cannot deny the similarity. These paradigms are presented as abstract models in the artificial semantic memory of computer linguistics. Let's consider this using the example of animals: animals are stored in the consciousness of native speakers of GLM of language as a community, essence, possibility. This group of lexical meanings in linguistics is called zoonyms, animal names such as wolf, bear, lion, horse, cow, cat, goat, deer, fox are artificially grouped in artificial semantic memory of computer linguistics under the national model [ Ott-Z00 ].
According to the linguist A. Sobirov, "when defining the paradigms of the semantic field, it is important to pay attention to the extensional and intensional features of things and events". According to L. Novikov, the extensional is a general class, and the intensional is a uniqueness corresponding to this class. Consequently, the extensional-intensional relation exactly corresponds to the dialectic of universality-particularity in the semantic field. For example, animals form a paradigm of zoonyms according to their extensional features, while the paradigms of wild animals and domestic animals differ according to their intensional features. In the artificial semantic memory of computer linguistics, zoonyms are included in the content of the national model [ ott-Z00 ] as a general class - extensional, and also as an internal paradigm, divided according to contradictory features into wild animals - in the content of the national model [Oty0^Qyj ]; domestic animals - [ Otf-;200 ]. Thus, the opposition of the members of a paradigm shows that a particular paradigm is the sum of several other paradigms.
Based on the approach outlined above, it is possible to model other paradigms in the semantic circle built on the basis of the category of extensional-intentionality of the dialectic of language and speech. For example, infinite hydronyms such as arroyo, branch, stream, river are given the extensional national model [ott-gldro ]; infinite phytonyms such as tree names, flower names, grass names ... - [Ott-flto ].
The famous linguist A. Nurmanov interprets paradigmatic relations as relations of units belonging to the same level, as a collection of similar concepts in certain nests. In his opinion, the general meaning underlying the unification of lexemes in one nest should be repeated in all
members of this nest. There are many such nests in the semantic field of Uzbek language. For example, general meaning in lexemes binafsha, ismaloq, momaqaymoq, otquloq, chuchmoma, yalpiz - plant; general meaning in lexemes qishloq, shahar, sahro - place; general meaning in lexemes bulbul, mayna, musicha, to'ti, qarchig'ay, qirg'iy - bird; general meaning in lexemes qalam, kitob, cho'mich, kosa, gilam, sholcha - item; general meaning in lexemes misqol, botmon, hovuch, siqim - quantity; general meaning in lexemes maktab, idora, do 'kon, bozor - organization, institution. Of these, the transposition model is created by moving the onomastic model in the form [Ott-fito ] for plant names that are not proper nouns; for geographical names that are not proper nouns - in the form [Ott-topo ]. Because in Uzbek onomastics proper names of plants, birds and places are combined into a separate name. However, there is a complexity of extensional and intensional modeling of the following lexical paradigms. Therefore, we limit ourselves to their intensional model: intentional national model in the form [ Ot^ashkilot ] is given for the paradigm representing organization-institution names; for a paradigm representing item names - [ Ot^uyum ]; for a paradigm representing quantity nouns - [ Ot{niqdor ]. This shows that it is necessary to define clearly the system of proper nouns.
The conflict of paradigms, built in the dialectic of language and speech, distinguishes a large set, a small set and classifications. This division exactly corresponds to our ideas about extensional-intensionality, based on A. Sobirov. In particular, models of semantic groups of nouns are built on the basis of the "set-classification" relationship. Therefore, within each large paradigm there are a number of small paradigms based on the conflict of certain features. For example, the set of zoonyms is divided into the following subsets: buzoq, tana, g'unajin, sigir, govmish, ho'kiz, buqa, ... cattle; ot, baytal, biya, ayg'ir, saman, yo'rg'a, yobu, to'riq, jiyron, bedov toy... horse; qo'y, qo'chqor, qo'zi, baqlan, hisori, arabi, merikos... sheep; it, qanjiq, ko'ppak, bo'ribosar, tozi, gurji, kuchuk... dog; eshak, mocha, xo'tik, xachir ... donkey.
Among the classifications (lexemes) of the above lexical sets, there is a difference in age, gender and species characteristics. In the artificial semantic memory of computer linguistics, lexemes are included in a separate model according to their characteristics. Sample: ot, baytal, biya, ayg'ir, saman, yo'rg'a, yobu, to'riq, jiyron, bedov toy ... included in the national model [ Ot*-200 ], in conflict according to gender characteristics the words baytal, biya, jiyron are included in the national model [ Ot*--°°) ]; the words ayg'ir, saman, yo'rg'a, yobu, to'riq, bedov - in the national model [ Ot„--°°) ].
As another example, we will consider a model of a set of cattle based on various contrasts: the national model of the set buzoq, tana, g'unajin, sigir, govmish, ho'kiz, buqa, ... is given in the form Otq-ra^j. If the members of this paradigm are sorted into the "chain of opposition"
according to their age, they are given in the national model [ Otq-,r°^ol ], the sign [ i...;U ] in the model represents the characteristic of growth in terms of age. If the members of the set are separated by gender, for tana, g'unajin, sigir, govmish the model [ ] is given; for
ho'kiz, buqa - the model [ Otq-ramoi^e) ], the signs [ U^ ] and [ i...;^ ] in these models expresses gender characteristics. If the age and gender characteristics of the members of the set are combined,
for tana, g'unajin, sigir, govmish the national model [ Otqoramoi^a) ]; for ho'kiz, buqa the national
"(a)
model [ Otq-r°m0l_(e) ] are given as a set-model.
In addition, linguists say that there are "equivalent contrast" between members of a paradigm. In this contrast, classifications (lexemes) act as a type of set. For example, bo 'rikalla, eskichopon, amiri, qirqma, gulobi, kampirqovun, umrboqi, aravakash, cho'gari are unified in melon; Qatra, felyeton, ocherk, novella, qissa, povest, roman - in prose; g'azal, tuyuq, ruboiy, fard, masnaviy, qasida, oq she 'r, poema, doston - in poem; drama, komediya, tragediya, tragikomediya - in drama; sho'rva, mastava, palov, ugra, lag'mon, manti, shovla, dimlama, qovurma - in meal and forms a set in the semantic field of the noun. These paradigms are stored as models in the artificial semantic memory of computer programs. As a source of semantic memory we presented the national model in the form [ OtqOVUn ] for the melon paradigm as invariant; for the prose paradigm - [ Ot„asr ]; for the poem paradigm - [ Ot„azm ]; for the drama paradigm - [ Ot^rama ]; for the meal paradigm - [ Ot£aom ]. These models are stored in the artificial semantic memory of computer linguistics as non-activated (not given to direct observation) communities, entities, possibilities, and sets.
A paradigm based on equivalence can move to the position of a small paradigm and form a large paradigm. For example, the small paradigms as prose, poem, drama are unified in (large) paradigm literature; the small paradigms sculpture, painting, cinema, theater are unified in paradigm art. In computer linguistics the national model in the form [ Otadabiyot ] is given for literature; [ Otsan,at ] - for art. The distinction between a large paradigm and a small paradigm or a large set and a small set in the semantic field of a noun leads to the conclusion that the paradigmatic relation is constructed according to a stepwise principle, in which the semantic field is assessed as the largest paradigm or the largest set.
Among paradigmatic relations, the relation of functionality, which is observed within the framework of a noun, is also of great importance. During direct observation, these speech units are generalized according to their function and their thought apparatus is created as an invariant model. For example, lexical units in the paradigmatic range o 'pka, bo 'g'iz bo 'shlig'i, unpaychalari, og'iz bo 'shlig'i, tanglay, til, tish, milk, lablar are concentrated in one place according to the function of forming speech sounds and included to the semantic memory of the national corpus of Uzbek language with the national model [ OtJ-^q ]; lexical units in the paradigmatic range tish, og'iz bo 'shlig'i, til, jig'ildon, oshqozon, o't, ichak, to 'qqizqorin, yo 'g'on ichak, buyrak are concentrated in one place according to the function of digestion and included to the semantic memory of the national corpus of Uzbek language with the national model [ OtJ-[m ]; lexical units in the paradigmatic range mashina, avtobus, trolleybus, tramvay, taksi, poyezd, qayiq, kema, paraxod, korabl, parom, aeroplan, samolyot, vertolyot, aerobus, raketa are concentrated in one place according to the function of transporting and included to the semantic memory of the national corpus of Uzbek language with the national model [ Otq-,tnov ]. However, depending on where the transporting takes place, small paradigms of functionality can be formed. In such situation, the words mashina, avtobus, trolleybus, tramvay, taksi, poyezd are given in the national model [ Otq-^OT-yer]; the words qayiq, kema, paraxod, korabl, parom - in the national model [ Otq-rtnov-suv ]; the words aeroplan, samolyot, vertolyot, aerobus, raketa - in the national model
[ ntl-v ]
L vyuqatnov-fazo ].
Based on the approach outlined above, countless paradigms are modeled, created naturally or artificially in the semantic field of a noun. This divides the semantic field into two parts - a feature, an event, a reality, given for direct observation, and the universality, essence, possibility
behind them. In computer linguistics, the functions performed by various automatic analyzers are a phenomenon, a reality given to direct observation, and the models that ensure the work of the automatic analyzer are an essence and a possibility, not directly observable. Everything is balanced by the law of materialization as an essence and its only phenomenon.
In another type of paradigmatic relations, as a result of the connection of members, a movement towards a common essence occurs, that is, the parts are collected and become a whole.
H. Nematov and R. Rasulov call this as incomplete contradictions. A. Sobirov evaluates it as a whole-piece relationship. In this paradigm, components are given as parts of a set, and if one of the parts is removed, incompleteness occurs. For example, kostum as a whole name requires the names of parts such as yeng, yoqa, tugma, yon cho 'ntak, ko 'krak cho 'ntak, о 'g'ri cho 'ntak, etak, jiyak, astar. In computer linguistics these parts are given in the whole model [ Otk_/um ]. Therefore, in the artificial semantic memory of computer linguistics, each whole is represented in a separate model and collects its parts. As a result, this type creates a special large-scale artificial semantic memory. In particular, the lexemes soch, qosh, qovoq, kiprik, ko'z, quloq, burun, og'iz, peshona, ensa, iyak are unified in the model [ Otbo/ ]; the lexemes barmoq, kaft, tirnoq, tirsak, yelka - in the model [ Otqo/ ]; the lexemes tirnoq, kaft, tovon, to'pqi, ilik, tizza, bolder, son - in
the model [ Otoy(/ ]. Also, these three paradigms, having moved to the position of a small paradigm, form the large paradigm bosh, oyoq, qo'l zanjiridagi and in computer linguistics the abstract model in the form [ Ot^,/ ] is created for them.
Also, the paradigmatic connection is realized in close connection with customs, traditions, socio-political life and history of native speakers. Therefore, it is necessary to take into account the linguocultural features of native speakers when modeling lexical semantic groups of nouns, generalizing the most important features of classifications, defining a set as the essence of this community. For example, if the national dishes such as sho'rva, mastava, palov, ugra, lag'mon, manti, shovla, dimlama, qovurma are included in the national model [ Ot£aom ], a separate model will be created for European dishes.
In general, the first, main criterion in the transmission of language and speech is "unobservability" and "observability". The relationship of paradigms in the semantic field of nouns is stored in the consciousness of native speakers as an entity that is not subject to direct observation. In the artificial semantic memory of computer linguistics, a model of such unobservables is collected as an entity, universality. However, the process of modeling is carried out by identifying, comparing, sorting and generalizing the "directly observable" feature, the most important features of the event. In this regard, N. Komlev says: "We see the world as it is, in its original state, and on this basis we divide and classify words into certain groups." Thus, when classifying objects and phenomena of the human world into classes, dividing them into types and genera, into certain semantic groups, first of all, we rely on the system of relationships between them, and this relationship is not violated in the process of modeling. What the linguo-cultural world of Uzbek speakers is, such a world should be artificially reflected in the national corpus of Uzbek language.
REFERENCES
I. Березин Ф.М. История лингвистических учений. - Москва: Висшая школа, 1975. - С.
50-51.
2. Неъматов X,., Расулов Р. Узбек тили систем лексикологияси асослари. Тошкент: Укитувчи, 1995 йил, 113-бет.
3. Никитин М.В. Основы лингвистической теории значения. - Москва: Высшая школа, 1988. - С. 26-27.
4. Новиков Л.А. Семантика русского языка. - Москва: Высшая школа, 1982. - С. 31-32.
5. Новое в зарубежной лингвистике. Вып.24. Компьютерная лингвистика. - Москва: Прогресс, 1989. - С. 10.
6. Собиров А. Узбек тили лексикасининг систем-структур тадкики. Монография. -Т.: "Booknomy print", 2022. - Б. 100.
7. Фалсафа (укув кулланма). -Тошкент: Шарк нашриёти-матбаа концерни бош тахририяти, 1999. - Б. 205.
8. Шемякин Ю.И. Начала компьчтерной лингвистики. Учеб пособие. - М.: Изд-во МГОУ, А/О «Росвузнаука», 1992.
