PECULIARITIES OF THE USAGE OF ARTIFICIAL INTELLIGENCE METHODS TO SOLVE DIGITAL TRANSLATION PROBLEMS IN THE MODERN LINGUISTICS (BASED ON THE UKRAINIAN-ENGLISH TRANSLATION)

Rita Kalko National Metallurgical Academy of Ukraine, by Street Vokzalny, 68 Sloviansk Ukraine 84109. ...................................................................................................................... Manuscript Info Abstract ......................... ........................................................................ Manuscript History Received: 05 April 2020 Final Accepted: 07 May 2020 Published: June 2020


Theoretical Review
The collective development of writing systems provides grounds for forming a working hypothesis about the possibility of deciphering the texts of related writing systems by referring to a common archaic source based on set theory. Cantor's "naive" set theory has already stated that any mathematical object is a set (Dauben, J., 1980). For example, a natural number is a set consisting of a single element of another set (a natural series), which, first of all, is itself a set of another order that satisfies Peano's axioms, and so on. Instead, Cantor's opponents, especially L. Kronecker, believed that only natural numbers and what is directly reduced to them could be considered mathematical objects.Such a narrowly specialized discussion between the proponents of Cantor and Kronecker is irrelevant to philology, since the studied sets of natural language alphabets are symbols of natural numbers by definition (Gowers, T.; Barrow-Green, J.; Leader I., 2008).
In the ancient alphabets, including Greek, Latin, Hebrew, etc., each letter also meant a number, and thus, by adding the numerical values of the letters, it was possible to obtain the numerical value of the whole name. This method was especially widely used by the Jews and was known as hematria ‫,‪)(Knohl‬גימטריה(‬ I. 2012). Such a phenomenon is well known in the European culture, beginning with the Revelation of John the Theologian about the "number of the beast": "He who has a mind, let him count the number of the beast, because it is a human number. And its number is six hundred and sixty-six (666 (ΥΞϚ))" (Chapter 13, verse 18).
According to many researchers, John the Theologian, describing the apocalyptic beast, meant the emperor Nero. It turns out that the inscription of Emperor Neuron in Jewish letters is obtained when counting 666. As religious scholar E. Duluman remarked, the number 616, found on many ancient sources, is due to the fact that the title of the emperor was written as "Nero Kesar", which, when counted, gave the number 616. The Greeks wrote the title of Nero as "Neron Kesar", which, when counted, gave the number 666 (Table 1). ‫ך‬  ‫ן‬  ‫נ‬  ‫ק‬  ‫ס‬  ‫ך‬  50  0  200  6  50  100  0  60  0  200  sum 666 It was then, during the course of two thousand years of history, the number of the beast was attributed to thousands of historical persons [6](DulumanY., 2009). The similar ambiguity of the attempt to analyze logically the Name as a set of phonemic and digital orders is better illustrated by a series of antinomies against the Cantor vulgar understanding of the set: the Burali-Forti paradox, Russell Grellinha-Nelson, Berry, Richard.
The indices do not coincide with other sets in set A: Я14, Я18, Я23, Я24 (Ξ, Φ, Χ); in the set B: b7, b17, b18, b22, b23 (G, Q, R, V, W); in the set C: c2, c4, c5, c7, c8, c10… c13, c16, c17, c19, c20, c24, c27, c33… c36, c39… c49  с2, с4, с5, с7, с8, с10…с13, с16, с17, с19, с20, с24, с27, с33…с36, с39…с49 (А, В, Ґ, Ѓ, Ђ, Ѐ, Є, Ё, Е, З, Ѝ, Ї, И, Љ, Њ, Ћ, Ќ, Т, Ѝ, Х, Ц, Ў, Ч, Ш, Щ, Ъ, Ы, Ь, Э, Ю) At the beginning, the most famous in the European tradition "digital" translation of the term "Neron Kesar" from Greek into Latin -"Nero Kesar" and numerous attempts to read it by different ethnic groups at different times were illustrated. This fragment is a part of the general Bible translations from the Aramaic globalized languages of the Roman Empire, the Greek Septuagint (completed in 132 BC) and the Latin Vulgate (Saint Jerome Ἱερώνσμος (340-420 AD)). The term is rewritten in Aramaic for the transfer of name and title, despite the affinity of the Greek and its descending Latin alphabet, then digitized on the principle of Letter-Number. The total sum of the numbers is equal to the ascending plus (or minus).Thus, this example has the specific characteristics of another language. The data clarify that the letter N and the number 50 characterize the flexibility of Latin.In our opinion, the orders of index values of the set of a certain alphabet selected in the work should be used in the translation practice according 215 to the principle -Litera-Index-Index value. It will be verified on the example of English translation (set D; index values d1, d2… d26) and in Ukrainian (E; e1, e2 … e33) (Table 8).Therefore, it is possible to distinguish index values of 3 ordersfor two sets D and E: high: DEd (digital match); average: DEf (phonemic coincidence); low D and E (no coincidence with another set) (table 4) Low (1) d10J1, d17Q1, d23W1, d25Y1 e2А1, e4В1, e5В1, e8Є0, e11З0, e13Ї0, e14И0 e27Х1, e28Ц1, e29Ч1, e30Ш1, e31Ы, e32Э0 e33Ю0 Sets D and E form subsets of different combinations of index values in the form of words, sentences and texts. The absence of a universal universal grammar for natural languages, despite numerous attempts to create it from the grammarians of Por-Royal (Grammaire générale et raisonnée de Port-Royal), allows us to perceive these subsets as a combination of "random" indices, which are subject to statistical (Gaussian) distribution, according to which the convergence of the distribution of the sum of independent equally distributed random variables approaches the normal distribution.
At first, let us do a simple experiment of digitalization of a subset of the word. At the same time, we shall consider only its index order. Concequently, the high order of the index has a numerical value of "3", medium "2" and low "1". In our study, it would be of special interest to use color universals.Traditionally, there are seven colors of the rainbow in the sequence of decreasing light wavelength: red (less than 740 nm), orange (orange), yellow, green, blue, blue, purple (more than 380 nm).Red 2 2 3. Церб-оний 1 2 2 2 2 2 1 1, etc. Referring to the example, it is possible to make the preliminary conclusion about performance of normal distribution on an example of separate words even on an index order.  . ed., 1963)Thus, we can also use the consonant-vowel criterion, which in the binary system is denoted by (1 and 0) (consonants are denoted by 1, because in the history of alphabetic writing the first characters were of the consonant type) (Red 1 0 1. Церб-оний 1 0 1 1 0 1 0 0, etc.).
Even a cursory glance allows us to draw conclusions about the presence in the Ukrainian translation of features characteristic of inflectional languages. This encourages the search for further common features.We combine the materials of the first and second stages from Table 8 (see Table 4). In general, the English alphabet shows the Gaussian distribution (1-4-0 for vowels (0) and 4-13-4 for consonants (1)), the Ukrainian alphabet shows 1-4-6 for vowels (0) and 4-10-7 by consonants (1)).
For further research, we change the index values within the sets of English (D) and Ukrainian languages (E). Traditionally, index values were denoted by a place in the alphabet (see table 8). The mentioned system had strong influences of external factors especially in the Ukrainian alphabet. In general, only from the time of writing I. Kotliarevsky's Aeneid in 1798 to 1905, up to 50 different spelling systems were proposed. Therefore, in our opinion, the applied set theory in the context of Gaussian (normal) distribution allows to change the index values of alphabets depending on the use of letters. Let us denote the index values of the letters in the new order using the order of the index from high to low (Table 5).
Noe we compare the first and second versions of digital translation in the Table 7. and final term is digitized taking into account the Index Procedure, Index Value and other indicators for translation adequacy assessment. This allows us to perceive the alphabetic symbols of European languages as mathematical sets, and the translation process itself can be explained not at the verbal but at the phonemic level, as a Gaussian distribution of random indicators.