For wordnet developing for a new language, the key problem is to find original resources that contain enough lexical data of the language in an appropriate format. This article discusses the structure, methodology of compilation and the current state of the bilingual Russian-Tatar Social-Political Thesaurus, which can serve as an initial resource for building the Tatar Wordnet. This thesaurus reflects the logical-semantic organization of lexical elements (synonymous, generic, and some other relationships) at the conceptual and lexical levels. Mainly, we focus on building synsets for nouns (single nouns and noun phrases).
A great hindrance to develop linguistic ontologies for a new language and conceptual modeling is the lack of original lexicographic resources containing full and relevant linguistic data description.
Success of Princeton WordNet has determined emergence of wordnets and
wordnet-like projects for different languages and multilingual wordnets. In wordnet
building developers often use the Expand Model
So the possibility of developing wordnets is largely determined by the presence of bilingual dictionaries or fairly complete descriptions of the semantic system of the language.
The absence of large English-Tatar dictionaries (the available ones are of very limited volume and may be used only for education purposes) makes it impossible to use the Expand Model to Tatar wordnet development, as well as absence of Tatar semantic dictionaries makes it almost impossible to develop original Tatar wordnet.
The objective of this article is to describe the methodology for constructing a Tatar wordnet based on a lexical resource such as the Tatar social-political thesaurus. This approach allows you to directly use the data of the thesaurus, primarily a set of synsets and relationships between synsets.
The body of the paper is organized as follows. Section 2 outlines the basic theoretical background of the study, and the main attention is paid to wordnet projects developed for the Turkic languages. Section 3 presents the methodology of compiling the Russian-Tatar socio-political thesaurus and its current state. Section 4 describes the most important aspects of implementing a wordnet-like resource using Tatar thesaurus synsets for Tatar nouns. Section 5 discusses the conclusions and outlines the prospects of future work. 2
At present time, there are various wordnets for some Turkic languages.
Two Turkish wordnet projects have been developed for the Turkish language. The
first one
Another Turkish wordnet is the KeNet
Unfortunately, the lack of large Turkish-Tatar dictionaries (as well as EnglishTatar ones) makes it impossible to translate Turkish resources into the Tatar language. In this respect the Tatar language can be attributed to low-resource languages.
The Extended Open Multilingual Wordnet
The BabelNet
Thus, the development of a qualitative Tatar wordnet with an emphasis on the specific features of the Tatar language based on the existing lexical resources is very relevant. 3
The conceptual model of the Tatar socio-political thesaurus (hereinafter referred to as
TatThes), the general principles of displaying linguistic data are taken from the
RuThes project (http://www.labinform.ru/pub/ruthes/)
In the RuThes each concept is linked with a set of language expressions (nouns, adjectives, verbs or multiword expressions of different structures – noun phrases and verb phrases) which refer to the concept in texts (lexical entries). The RuThes concepts have no internal structure as attributes (frame elements), so concept properties are described only by means of relations with other concepts.
Each of the RuThes concept is represented as a set of synonyms or near-synonyms (plesionyms). The RuThes developers use a weaker term, ontological synonyms, to designate words belonging to different parts of speech (like stabilization, to stabilize), the items may be related to different styles and genres. Ontological synonyms are the most appropriate means to represent cross-linguistic equivalents (correspondences), because such approach allows us to fix units of the same meaning disregarding surface grammatical differences between them. For example, Table 1 represents basic ways of translating Russian adjective + noun phrases into Tatar.
The TatThes is based on the list of concepts of the RuThes, i.e. the Tatar component is based on the list of concepts of the RuThes thesaurus. The methodology of compiling the Tatar part of the thesaurus includes the following steps: 1. Search for equivalents (corresponding words and multiword expressions) which are actually used in Tatar as translations of Russian items. 2. Adding new concepts representing topics which are important for the sociopolitical and cultural life of the Tatar society and which are not presented in the original RuThes (for example, Islam-related concepts, designations of Tatar culture specific phenomena, etc.). 3. Revising relations between the concepts considering the place of each new concept in the hierarchy of the existing ones and, if necessary, adding the new concepts of the intermediate level. So an important step is to check up the parallelism of conceptual structures between the languages.
The TatThes is mainly being compiled by manual translation of terms from the
RuThes into Tatar, besides the Tatar language specific concepts and their lexical
entries are added (about 250 new concepts). Search for equivalents in the Tatar language
in many cases became a time-consuming task, because available Russian-Tatar
dictionaries of general purpose contain obsolete lexical data
In the course of the project we found that distinguishing feature of the
contemporary Tatar lexicon is a great deal of absolute synonyms of different origin in and
structure, the main cause of the phenomenon is language contacts
The TatThes is implemented as a web application and has a special site
(http://tattez.turklang.tatar/). Additionally, it has been published in the Linguistic
Linked Open Data cloud as part of RuThes Cloud project
Previously, the RuThes thesaurus has been semi-automatically converted to the
WordNet-like structure, and Russian wordnet (RuWordNet) has been generated
The current version of RuWordNet (http://ruwordnet.ru/eng) contains 110 thousand Russian unique words and expressions. The same approach can be used to transform TatThes to Tatar wordnet.
The TatThes data may be serve as an initial basis for wordnet building by the following reasons: 1. The sociopolitical sphere covers a broad area of modern social relations. This area comprises generally known terms of politics, international relations, economics and finance, technology, industrial production, warfare, art, religion, sports, etc. 2. Currently the TatThes, in addition to terminology, comprises some general lexicon branches representing lexical items which can be found in various domain specific texts. 3. Semantic relations in the TatThes are necessary and sufficient to arrange the Tatar nominal vocabulary (nouns and noun phrases) as a wordnet-like network of synsets.
Thesaurus concepts unite synonymous items, so we have ready sets of synonyms as building blocks for wordnet. The concepts are linked by semantic relations with each other. In the RuThes and in the TatThes there are four main types of relationships between concepts, see Table 2. Semantic relations, mapped in wordnet, are not all shared by all lexical categories, so thesaurus data converting into wordnet format require dissimilar ways for different parts of speech. Asc and Asc1/Asc2 association relations need additional explanations. The Asc symmetrical association, distinguished in RuThes and inherited by Tatar Socio-Political Thesaurus, connects very similar concepts, which the developers did not dare to combine into the same concept (for example, cases of presynonymy of items).
The Asc1/Asc2 asymmetric association connects two concepts that cannot be described by the relations mentioned above, but neither of them could not exist without the existence of the other (for example, a concept SUMMIT MEETING needs existing the concept HEAD OF THE STATE). In studies of ontologies this relation may be mapped as the ontological dependence relation.
Nevertheless, basic semantic relations which we need to group nouns concepts into wordnet are presented in the TatThes.
The core of the TatThes is made up of nouns and noun phrases (see Table 3), so the bulk of thesaurus data may be used for Tatar wordnet building without significant changes (synonymous items are yet joined into synsets and the required relations between them are selected).
An important issue is reflecting Tatar language specific word usage features in the resource. Presence alone of the shared concepts in languages do not necessarily evidences the same ways of usage of individual words or of usage words of individual semantic classes. Consider this with an example. Specific feature of the Tatar language is using of hypernyms before a corresponding hyponym, and such using is not regarded as pleonasm in many cases (examples 1–3): (1) Париж шәһәрендә ‘in the city of Paris’ (instead of ‘in Paris’); (2) кыз кеше ‘girl human’ (instead of ‘a girl’); (3) май аенда ‘in the month of May’ (instead of ‘in May’).
In cases when such a usage is conventionalized and corpus data evidences that the usage has a high frequency, we include such hyponym-hypernym items into a list of lexical entries of a concept. Such manner of designating is a feature of using toponyms and some classes of general lexicon, so it should be considered in Tatar wordnet building. For example, lexical entries of month names include such conventionalized noun phrases, composed of the month name and the hyponym, designating month in general, see Table 4.
Because the RuThes concepts assemble ontological synonyms, the RuThes lexical entries bring together words of different part of speech. Therefore, in standard case a Russian synset joins a noun (often we use it as a concept name) and a relative adjective derived from a noun (Table 5; only core items of synsets are represented). In Tatar, like in other Turkic languages, there is no original relative adjectives (and existing ones are borrowed from European or Oriental languages), so in many cases the TatThes synsets are composed of items of the same part of speech, mainly of nouns. This circumstance greatly facilitates cleaning thesaurus synsets data for wordnet developing.
Basic lexical entries of a Russian Part of speech of Basic lexical entries of Part of speech of concept Russian words a Tatar concept Tatar words Река ‘river’ N Елга ‘river’ N Речной ‘of river, fluvial’ ADJ Факультет ‘faculty’ N Факультет ‘faculty’ N Факультетский ‘of faculty’ ADJ Преподаватель ‘teacher N Укытучы ‘teacher’ N Преподавательский ‘of teacher’ ADJ Больница ‘hospital’ N Хастаханә ‘hospital’ N Больничный ‘of hospital’ ADJ Сырхауханә ‘hospital’ N
So the core of the TatThes is made up of nouns and noun phrases (69% of total number of concepts). At the moment semantic relations between nouns mapped in thesaurus, are necessary and sufficient to convert Tatar thesaurus data into the wordnet format.
4
When building a wordnet for a new language, in particular, for a low-resource one, a crucial issue is searching for appropriate sources. We are planning to use data of the TatThes as a base resource for developing Tatar wordnet.
The TatThes is being compiled by manual translation of terms from the RuThes into Tatar, with searching Tatar equivalents used in real texts, so the thesaurus contains relevant lexical data. In the TatThes each concept is linked with a set of language expressions (single words or multiword expressions) which refer to the concept in texts – lexical entries.
The analysis of thesaurus data shows that the bulk of the thesaurus synsets are formed around nouns or noun phrases. A mapping semantic relations of nouns in the thesaurus reproduces a mapping semantic relations in wordnets.
Future work includes adding material of verbs and other parts of speech. Also we are planning to develop some automatic approaches to mining terms and to asses the Tatar terminology coverage in Thesaurus on Tatar socio-political texts data.
This work was partially funded by the subsidy allocated to Kazan Federal University for the state assignment in the sphere of scientific activities, grant agreement no. 1.2368.2017 and partially by a subsidy assigned to the Institute of Applied Semiotics of the Tatarstan Academy of Sciences for the state assignment.