Giorgi Chikoidze

The main problem of language modelling is the meaning of the utterance, which is the ultimate goal of one of the directions of language model functioning (analysis) and the starting point for its opposite direction (synthesis). The paper suggests one of the options of the problem solution. In particular, dividing sentences into the set of (quasi)-synonymous “sentential primitives" and on the basis of the "primitives", building such a structure, which reflects the appropriate content of this set quite transparently. Determination of the estimated set of primitives should be based on the search of the constituents according to their characteristics, particularly, to their syntactic properties. Verbs and their dominant role in sentences are considered according to "multilevel" syntax. By this theory, a primitive is a central component (core) of the layered structure, which is "surrounded" by the periphery i.e. traditional adverbs. The central component itself involves the verb and its actants. At the same time, internal relations of primitives, i.e. dependencies of actants on the predicates and peripheries on central structures are defined by the semantic roles. Such approach to language content conveys the meaning of the utterance through the sentential primitives where the primitives are the language units. In the frame of the presentation, we will offer a translation model, which is simplified to the level of primitives. For testing the model, Georgian-English parallel corpus will be used.

Fundamental scientific researches in Computational Linguistics have been carried out at the Archil Eliashvili Institute of Control Systems, Georgian Technical University, for many years. Various combination methods (lexical functions, synonymous series, semantic roles and superparadigms) were developed for the Georgian language; A computer dictionary of the Georgian language has been created, which, at the same time, carries out functions of a morphological generator, in other words, it produces the full paradigm for each lexical unit. Within the framework of the fundamental issues of language modeling, a means of presenting linguistic algorithms has been developed, that allows the formulation of a bi-directional analysis-combined processor. For some languages, the filling-widening process of dictionaries ​​has been simplified with the help of a grammar compiler, which is the most modern tool for the automatic realization of a formal language model. It is possible to compile morphological processor libraries of individual languages ​​for different variations of any language (according to time, space, origin, genre, etc.) and so on. Automatic machine translation can be considered as the main achievement. To solve this task completely it is necessary to create lexical translator. This type of system is rather valued among the ordinary users, as it makes it easier for them to learn foreign-language texts intensively thus is much more useful while composing text.The strategy of our team is to provide reliable support for future language technologies by the theoretical and practical key issues that have been worked up in separate projects for the past years. The computer products created by our team are used in various linguistic areas. It is a challenge for linguists to create computational model of a language, taking into account its multi detections and changes. We have created main components to compile a national corpus manager.

The definition of a Georgian sentence role structure is based on the nation of verb super-paradigm, which represents the unit of paradigms generated from a single lexical unite. These paradigms are ordered as an action chain, where each element influences its immediate right “neighbor”. Basic participants of this global process are causer, agent object and addressee. Marking of the text corpus by this information makes more clear the semantic structure of the sentence, what, in its turn, makes such corpus more useful for creation and application of the language models.

The paper is dedicated to the discussion of the selection scheme of Georgian sentence lexemes. The scheme organizes and regulates the selection of lexemes that fill up the role structure of the sentences. The proposed scheme is based on the concepts of semantic roles and lexical functions.

The analysis of a sentence represents one of the directions of the meaning↔text model functioning. As different from the synthetic direction, the initial point of the process of analysis is represented by a graphical representation subordinated to immediate perception. Specifically, a text is shown as a coordination of symbols where textual word forms are represented in a chain separated by space bars, some of them being highlighted by specific symbols or punctuation marks. Such a sharp separation of wordforms first of all simplifies the analysis of a text in comparison with speech in which continuous chains of phonemes often represent a group of some words and not separate wordform. Such kind of a peculiarity provides an opportunity for an immediate and unequivocal identification of each textual word form with a lexical unit, followed by certain grammatical and semantic characteristics. Morphological annotation is a rather important aspect of analysis, provides a basis for furthering of the process of analysis to its ideal (yet not achieved) aim or immediate representation of „meaning“. The next important step is to establish the whole structure of a sentence on the basis of already procured morphologic information which is characteristic to separate elements (word forms). Structure building and representation is made differently (for instance, by the graph of immediate constituents).

According to Melchuk, the left side of the “Meaning↔Text” language model should be represented by a set of synonymic utterances and some most simple, “neutral” member of it can be considered as a “dominant” of this set, which represents a semantic core of the set and correspondingly of the “Meaning”. Proceeding from this, the basis of pragmatic-semantic level functioning should be implemented by quasi-synonymic transfigurations, which lead from the dominant to the arbitrary utterance of synonymic set (during synthesis) or in an opposite direction (in the case of analysis), taking into account in the course of these operations the pragmatic-semantic nuances, which make difference between initial and final arguments of each transformation. The possibility and result of these transformations are defined by same special information, which should be added to the morphologic and syntactic data of dictionary entities. We shall mention three kinds of such information. The first of them is represented by the spectrum of “lexical functions”, which can be considered as some version of the lexical paradigm: they define for each lexeme set of content in which it acquires one of semantic nuances, defined by same concrete lexical function. The “Synonymic rows” can be considered as a generalization of one of the lexical functions, particularly, of Syn, which implies the unity of words synonymic to the lexeme of the given dictionary unit. The concept of “synonymic rows” implies more “base” conditions of quasi-synonymy, which at the same time must be followed by description of all features, which define the semantic differences between these quasi-synonyms. The Georgian verbal super-paradigms, in their turn, may be considered as a particular case of quasisynonymic rows: these entities unite all paradigms derived from the one and the same lexeme. The main advantage of such unification is the stability of possibilities and results of transformations between utterances based on the different members of superparadigm. 

The report discusses some components of the range of linguistic means that give the expression desired (sometimes even necessary) communicative tinge: it adds a sort of general or specific (quasi-synonymous sequence) semantic nuance to each member, or accentuates this or that member (negative - in case of skipping this member, which allows the super-paradigm, or "positive" - ​​by changing the order of members).

Lexical functions LF () may be considered as something similar to the lexical “paradigm” of the corresponding vocabulary unit C0; which includes stable lexical units associated with C0 and connected to it by some meaningful and standard semantic relations. Particularly, the inclusion of lexical functions both in the dictionary units and in the system of a language is that this information often supplies the possibility of correct choice of a word combination in the text: gasca brÂaneba - ordered; gaak’eta gancxadeba - announced, miiγo gadac’qvet’ileba - decided; misca rčeva - advised;… (1). The Oper1 (C0), represented by verbs of (1), cannot be replaced by each other in spite of the semantic likeness of their arguments, and thus, the correct choice of these verbs fully depends on the corresponding C0. On the other hand, the components of lexical functions (LF) may serve as one of the tools for quasi-synonymous transformations.

The lion share of mental (philosophical, logical) concepts has its origin in the natural language - “body of thinking”. The most important of those concepts should have their origin in all languages, though with some possible additional nuances which make them more understandable and near for usual users of this language. Particularly, here will be given a short review of Georgian conjunctions with semantics corresponding to the logical implication.

Algorithmic language description of linguistic in new and more important sphere in this case, not only to become comparison static system, As a matter of operating mechanism, which in turn receives and creates a text (speech) of his functions in the process. Such as it can be described and the language to reflect the main activity. The article shows the view on the basis of the models through the translation industry. 

Approach to the modeling of language knowledge actuation, given in the paper, is based in first instance on the widely accepted statement that just the generative grammar represents a model of the language knowledge. Proceeding from this the scheme of morphologic generator correction, which is a main topic of the work, may be considered as a try to imitate in the frames of an interactive regime some aspects of the language teaching/learning process. The earlier elaborated Russian morphologic generator serves as an immediate concrete object of this pilot scheme.

Our research is based on an understanding of natural language meaning that attaches crucial importance to the share of spatial relations. According to this approach the content of a sentence is conceived as a projection in the Language Space (LS), a somewhat transformed reflection of the real space. Projections are thus one component of LS, the other being metaphorical reflections of quality degrees, functioning of the perception organs, social relations, etc. The characteristic features of LS are the such : position of the observer, the focus of her/his attention, direction of their movements (back, forward, up, down, etc.). The most important role is attached to the border, its position and orientation relative to the observer and the object of his/her attention and their movements. Besides the prepositions and verb roots a paramount contribution to the organization of LS projections is due to verbal prefixes, at least for languages with highly developed verb prefixation (Russian, German, Georgian, etc.) In our present work we deal with schematizing the LS of the two most common Russian verbal prefixes: v– (‘in’) and vy– (‘out’). A comparative perspective on the problem is supported by parallel examples from German and Georgian. In the conclusion, the interpretation of the meaning of verbal prefixes is brought into correlation with metaphorical expressions.

The correspondence between event participants profiled by Georgian verb and their case markers is considered. As a basis for explanation of these relations is proposed the metaphor of “Action chain/Energy flow” (R. Langacker 1991). It’s supposed that option of the case of nominal expression for participant depends on its position within the action chain”: in first instance, is defined by initial and final points of this linear structure and then by some additional factors

Bfp is morphological generative system (Basic form→Paradigm), which generates for each input basic form (Bf) its whole paradigm (P). As yet it is implemented and tested on the Russian verb paradigms only. However this system is already connected with the Printing Support System (PSS) and is supposed to represent its morphological component. According to PSS restrictions it will demonstrate on the screen in consecutive order subsets of verb paradigms including less than 10 members. The hypothesis is postulated that NL mechanism may function according to the somewhat alike scheme

A co-operative project between two research groups in Tbilisi and Stockholm began in 1996. Its purpose is to extend a word predictor developed by the Swedish partner to the Russian language. Since Russian is much richer in morphological forms than the 7 languages previously worked with, an additional morphological component, using an algorithm supplied by the group in Tbilisi, is seen as necessary. It will provide inflectional categories and resulting inflections for verbs, nouns and adjectives. The correct word forms can then be presented to the user of the word prediction system in a consistent manner, allowing the user to easily choose the desired inflectional word form. At present, the work with the classification of verbs is complete. The algorithm is also being used to automatically tag the large lexicon used in the word predictor with inflectional classes.

This work is aimed at definition of Georgian genitive’s meaning. Some contexts of its use are considered and on the basis of their similarities and differences a scheme is constructed which supposedly should subsume the main features of their content. The crucial point of this approach is a concept of object “environment”, sometimes named – Influence Area (IA); the prototypical meaning of Georgian genitive case can be described as “attraction” of some object O2, external to the one marked by genitive (O1), to the “environment” of the latter. This prototypical version of scheme subsumes all cases of the most usual possessive construction (Ng+N), however some important details depend on the lexical features of its components (Ng, N). It turned out that these lexical restrictions to be satisfied require subdivision of the object “environment” in some subareas each of which has its peculiar semantical relation to the core-object.

Principal scheme of a system, generating word paradigm proceeding from its basic (dictionary) form, is proposed. The system is represented by morphologic nets, which were earlier successfully used for bi-directional (synthesis/analysis) morphologic processors and for morphologic tagger without dictionary. The system under consideration uses instead of full range dictionary list of basic forms, which is structured in the way, that belonging to some of its sub-lists is sufficient for definition of all characteristics necessary for building of corresponding paradigm. The system is tested on Russian verb paradigms. Being accomplished it should serve as automatic Russian spelling dictionary.

Doctoral Thesis Referee

Master Theses Supervisor

Doctoral Thesis Supervisor/Co-supervisor

Scientific editor of monographs in foreign languages

Scientific editor of a monograph in Georgian

Editor-in-Chief of a peer-reviewed or professional journal / proceedings

Review of a scientific professional journal / proceedings

Member of the editorial board of a peer-reviewed scientific or professional journal / proceedings

Participation in a project / grant funded by an international organization

Participation in a project / grant funded from the state budget

Patent authorship

Membership of the Georgian National Academy of Science or Georgian Academy of Agricultural Sciences

Membership of an international professional organization

Membership of the Conference Organizing / Program Committee

National Award / Sectoral Award, Order, Medal, etc.

Honorary title

Monograph

Handbook

Research articles in high impact factor and local Scientific Journals

Publication in Scientific Conference Proceedings Indexed in Web of Science and Scopus