Computational linguistics and intelligent systems
Permanent URI for this communityhttps://ena.lpnu.ua/handle/ntb/39447
Browse
Item Methods of comparing interval objects in intelligent computer systems(National Technical University «KhPI», 2017) Shepelev, Gennady; Khairova, Nina; Institute for Systems Studies of Federal Research Center ―Computer Science and Control of Russian Academy of Sciences, Moscow, RussiaProblems of expert knowledge representation by means of generalized interval estimates approach and using methods of comparing interval alternatives in the framework of intelligent computer systems are considered. The problems are common in economy, engineering and in other domains. Necessity of multi criteria approach to comparing problem that is taking into account both preference criteria and risk ones is shown. It is proposed to use a multi-steps approach to decision-making concerning choice of preferable interval alternatives. It is based on consistent using of different comparing methods: new collective risk estimating techniques, ―mean-risk‖ approach (for interval-probability situations) and Savage method (for full uncertainty situations).Item Титульний аркуш до "Computational linguistics andintelligent systems (COLINS 2017)"(National Technical University «KhPI», 2017)Item Unsupervised acquisition of morphological resources for Ukrainian(National Technical University «KhPI», 2017) Hamon, Thierry; Grabar, Natalia; LIMSI-CNRS, Orsay, Université Paris 13, Sorbonne Paris Cité, France; CNRS UMR 8163 STL, Université Lille 3, 59653 Villeneuve d'Ascq, FranceAvailability of morphological resources is an important and recurrent need because they allow the development of NLP tools and applications for a given language. Indeed, such resources provide basic information which is necessary for such tools for performing more sophisticated treatments (information retrieval, morphosyntactic tagging, etc). We propose to acquire morphological resources for Ukrainian language. The method proposed exploits corpora in order to extract words that are related morphologically between them. The method has two versions: without and with processing of prefixes. The association strength between these words indicates their probability to have a morphological and semantic relation between them. We use three corpora (literary, medical and general-language) and evaluate the results obtained. According to the corpora, precision varies between 67% and 86%. The results from different corpora are also compared, which shows that there is little redundancy between the corpora. The currently available resource contains 3,315 fully validated pairs of words.Item Use of linguistic criteria for estimating of wikipedia articles quality(National Technical University «KhPI», 2017) Kolesnik, Anastasiia; Khairova, Nina; National Technical University "Kharkiv Polytechnic Institute"Item Improving communication in enterprise solutions: challenges and opportunities(National Technical University «KhPI», 2017) Gorbachov, Vitaliy; Cherednichenko, Olga; National Technical University "Kharkiv Polytechnic Institute"Item Analysis of existing German Corpora(National Technical University «KhPI», 2017) Olifenko, Inna; Borysova, NataliaItem Semantic state superpositions and their treatment in virtual lexicographic laboratory for spanish language dictionary(National Technical University «KhPI», 2017) Kuprianov, Yevgen; Ukrainian Lingua-Information Fund, NAS of UkraineThe paper is devoted to ambiguities of Spanish language units: their formal modelling and treatment in the virtual lexicographic laboratory VLL DLE 23. The final goal is to find optimum solution for lexicographic treatment and research of ambiguities in the laboratory. As a theoretical base for developing ambiguity model, the theory of semantic states was selected. The ambiguity, i.e. the acquisition of different meanings by the unit at the same time in a given context, is represented in the model as a superposition of respective semantic states.Based on literature materials, the formal model of superpositions describing ambiguity formation mechanism in Spanish units was built. The model was further used to make out the interface intended for treating semantic state superpositions in VLL DLE 23.Item Methods and models of automatic ontology construction for specialized domains (case of the Radiation Security)(National Technical University «KhPI», 2017) Orobinska, Olena; Chauchat, Jean-Hugues; Sharonova, Natalya; National Technical University "Kharkiv Polytechnic Institute"We propose a hybrid, semi-automatic approach that uses the intersection of semantic classes of nouns and verbs built on the domain lexicon and builds kernel ontology from a list of initial concepts and then completes this kernel ontology by new entities detected in a large corpus of texts of international standards of Radiological Safety. The results confirm the important role of initial linguistic modeling and show that the external lexical resources available online can contribute effectively to the resolution of the problem of lexical disambiguation.Item Evaluation ofa formalized model for classification of emergency situations(National Technical University «KhPI», 2017) Titova, Vera; Gnatchuk, Ielizaveta; Khmelnitsky National UniversityFormalization of conditions that characterize the problem of classification of emergency situations is considered in this paper.This formalization is the basis for the Formalized Model of the emergency situations classificationproblem. Intelligent methods are used to solve this problem. These methods are also the basis for the development of the Neural Network Model for emergency situation classification. In this paper wedevelop the structure of the model and determine the number of network layers, the types of neurons and its membership functions. Using the Neural Network Model as decision support for the dispatchers of emergency services makes it possible to improve the quality of emergency situations classification.Item Creation of a multilingual aligned corpus with Ukrainian as the target language and its exploitation(National Technical University «KhPI», 2017) Grabar, Natalia; Hamon, Thierry; CNRS UMR 8163 STL, Université Lille 3, 59653 Villeneuve d'Ascq, France; LIMSI-CNRS, Orsay, Université Paris 13, Sorbonne Paris Cité, FranceThe question on creation of linguistic resources (such as corpora, lexica or terminologies) occupies an important place in the research areas related to linguistics, Natural Language Processing, Computer Sciences, psycholinguistics, etc. In this paper, we propose the description of a multilingual corpus in which Ukrainian is the target language, while source languages are Polish, French and English. The corpus contains literary texts and a small subset built with texts provided by medical area. On the whole, the corpus is composed of 62 literary texts and 129 medical texts. The corpus counts over 1 million words in the tar-get Ukrainian language, and at least as much in the source languages taken all together. This is a directional corpus aligned at the level of sentences. After the description of this corpus, we introduce some possible exploitations and first results. We then conclude and indicate some directions for future work. The corpus presented in this work is available for the research purposes: http://natalia.grabar.free.fr/resources.php.Item Gamification: today and tomorrow(National Technical University «KhPI», 2017) Yukhno, Katherine; Chubar, Eugenia; National Aerospace University ―Kharkiv Aviation InstituteItem Discursive units in scientific texts(National Technical University «KhPI», 2017) Verbinenko, Yulia; Ukrainian Lingua-Information Fund of NAS of UkraineDiscursive units are text elements that ensure its coherence, direct attention to the context, make text clear etc. Undeveloped theory of semantic description and its lexicographical representation complicates the description of the discursive units. There are also difficulties in dictionary definitions formulating, as discursive units are often very integrated into the context. Because of this, it is difficult to define system boundaries and build up the correct classification. The main criterion for merging of heterogeneous units into one class of discourse units is their joint function of regulation and organization of the communication process. It is impossible to classify discursive units only by grammatical (morphological and syntactic) features. In terms of morphology, these units are also difficult to combine into one class. In our opinion, it is functional feature that is the most relevant for determining discursive units in the text. Therefore, semantic-pragmatic characteristics are most relevant for the determination of the discursive units in the text.Item Search optimization and localization of the website of Department of Applied Linguistics(National Technical University «KhPI», 2017) Pidpruzhnikov, Vsevolod; Ilchenko, Margarita; National Aerospace University ―Kharkiv Aviation InstituteItem Intelligent data processing in creating targeted advertising(National Technical University «KhPI», 2017) Kirkin, Stanislav; Melnyk, Karina; National Technical University "Kharkiv Polytechnic Institute"Item Intelligent system structure for Web resources processing and analysis(National Technical University «KhPI», 2017) Lytvyn, Vasyl; Vysotska, Victoria; Chyrun, Lyubomyr; Smolarz, Andrzej; Naum, Oleh; Lviv Polytechnic National University; Institute of Electronics and Information Technology, Lublin University of Technology; Information Systems and Technologies Department, Drohobych Ivan Franko State Pedagogical UniversityThe paper describes the general detailed and formal description of intelligent system of information resources processing (ISIRP) based ontology. The content life cycle phase implementation of ISIRP structure is improved. The general principles of ISIRP designing structures enable automated information resource processing to increase regular user text content realization, reducing the production cycle, saving time and increasing the e-commerce capabilities.Item Зміст до "Computational linguistics andintelligent systems (COLINS 2017)"(National Technical University «KhPI», 2017)Item Statistical methods usage of descriptive statistics in corpus linguistic(National Technical University «KhPI», 2017) Didusov, Valeriy; Kochueva, Zoia; National Technical University "Kharkiv Polytechnic Institute"Item NLP resources for a rare language morphological analyzer: danish case(National Technical University «KhPI», 2017) Kotov, Mykhailo; V.N. Karazin Kharkiv National University, Kharkiv, UkraineThe paper discusses the characteristics and practical aspects of application of the natural language processing resources available for developing a rare language morphological analysis solution. The case under consideration reveals the pipeline design needed to prepare the grammatical resources for Danish. Being rare not only in terms of distribution, but also in the amount of natural language resources available, the Danish language represents a significant problem in terms of application of third-party tools to help solve various NLP-related issues. The paper focuses on part-of-speech tagging and lemmatization, typical but indispensable tasks at the pre-processing stage within the framework of developing a morphological analyzer as a custom NLP solution.Item An index of authors’ popularity for Internet encyclopedia(National Technical University «KhPI», 2017) Lande, Dmitry; Andrushchenko, Valentyna; Balagura, Iryna; Institute for information Recording of NAS of Ukraine, KyivThe new index of the author‘s popularity estimation is represented in the paper. The index is calculated on the basis of Wikipedia encyclopedia analysis (Wiki-Index–WI). Unlike the conventional existed citation indices, the suggested mark allows to evaluate not only the popularity of the author, as it can be done by means of calculating the general citation number or by the Hirsch index, which is often used to measure the author‘s research rate. The index gives an opportunity to estimate the author‘s popularity, his/her influence within the sought-after area ―knowledge area‖ in the Internet – in the Wikipedia. There are proposed algorithms and the technique of the Wiki-Index calculation through the network encyclopedia sounding, the exemplified calculations of the index for the prominent researchers, and also the methods of the information networks formation – models of the subject domains by the automatic monitoring and networks information reference resources analysis.Item Content analysis of some social media of the occupied territories of Ukraine(National Technical University «KhPI», 2017) Lytvynenko, Volodymyr; Lurie, Iryna; Radetska, Svitlana; Voronenko, Mariia; Kornilovska, Natalia; Partenjucha, Daria; Informatic and Computer Science Department, Kherson National Technical UnivesityThe paper analyzes the activities of the Internet publications in the temporarily occupied territories. The analysis of the tools for the detection and monitoring of free and paid services, data analysis of social media to monitor references and comments on social networks are presented in the paper. This paper includes an example of a practical text analysis and extraction of information online. It is shown that the use of computing environment KNIME opens new pathways in a variety of social media and determine the user behavior. This technique improves the process of analytical studies.