English
DonNTU  Portal of Masters

Library of materials on the topic of graduation work

    Own publications and reports

  1. APPLICATION OF MACHINE LEARNING ALGORITHMS FOR DETERMINING THE TONALITY OF THE TEXT OF REVIEWS ON THE INTERNET STORES

    Authors: N.A.Gorbunov , I.A.Kolomoitseva

    Description: In this article we will consider the use of machine learning algorithms to determine the tonality of text reviews in online stores. We will also look at examples of the use of such algorithms and possible ways to improve the effectiveness of these solutions. .

    Source: Collection of materials of the student section of the XIV International Scientific and Technical Conference.

  2. DYNAMIC CALCULATION OF A SOLID-STATE PART USING THE METHOD FINITE ELEMENTS

    Authors: N.A.Gorbunov, A.V. Grigoriev.

    Description: In this article, the procedure for the practical application of the finite element method is considered, the advantages of the method and its features are described, the method of structural analysis by the finite element method is considered. The method of direct equilibrium, methods of work or energy, the principle of virtual work, methods of weighted balances, computer methods of solving problems were touched upon. Conclusions are drawn. .

    Source: Collection of materials of the student section "Methods and means of automation of software and systems

  3. FEATURES OF FURNITURE DESIGN IN CAD, TECHNOLOGIES AND TECHNIQUES. FINITE ELEMENTS

    Authors: N.A.Gorbunov, A.P.Ishchenko, A.V. Grigoriev

    Description: The article considers the problem of implementation and design using CAD in the enterprise, the pros and cons of automation, features of furniture design, enumeration of advantages and disadvantages of furniture modeling programs, business development prospects .

    Source: Collection of articles of the student collection

Feature articles

  1. ANALYSIS OF THE TONALITY OF TEXT REVIEWS USING TONAL DICTIONARIES AND CARDINALITY OF A FUZZY SET

    Authors: E.M.Gerasimenko, V.V.Stetsenko

    Description: This article describes one of the most common methods used in analyzing the tonality of text content - natural language processing

    Source: Journal: Proceedings of the Southern Federal University. Technical sciences

  2. INTELLIGENT TONALITY ANALYSIS SYSTEM FOR MEASURING CUSTOMER LOYALTY AND DECISION-MAKING BASED ON FUZZY LOGIC

    Authors: E.M.Gerasimenko, V.V.Stetsenko

    Description: An intelligent approach to measuring customer loyalty to a specific product based on the analysis of comments is presented. A general analysis of the tonality in tweets and messages is carried out quite often, but task-oriented analysis of users' opinions and measuring their loyalty level is a new idea.

    Source: Journal: Proceedings of the Southern Federal University. Technical sciences

  3. FORECASTING CUSTOMER LOYALTY TO BROADBAND INTERNET SERVICE USING FUZZY LOGIC ALGORITHMS

    Authors: R.N.Ermakov

    Description: The methodology of constructing logical-linguistic models for solving the problem of predicting customer loyalty to a service (broadband Internet access) depending on the quality of services provided in fuzzy conditions is considered. For the practical application of the methodology, software has been developed using fuzzy logic algorithms.

    Source: Journal: Economics and Quality of Communication Systems 2019

  4. Deep Learning

    Authors: I.Goodfellow, Y.Bengio, A.Courville

    Description: This book contains an extensive overview of deep learning topics, including the basics of neural networks.

    Source: Book,electronic version

  5. Natural Language Processing in Action

    Authors: Y.Goldberg

    Description: This book covers the field of natural language processing (NLP), which is important for analyzing the tonality of text.

    Source: Book,electronic version

  6. Neural Network Methods in Natural Language Processing

    Authors: Lane, Howard, and Hapke

    Description: This book is more narrowly focused on the use of neural networks in natural language processing, which includes tonality analysis.

    Source: Book,electronic version

Translated article

  1. Text Tonality Classification Using a Hybrid Convolutional Neural Network with Parallel and Sequential Connections Between Layers

    Authors: R.Peleshchak1, V.Litvin1, I.Peleshchak1, A.Khudy, Z.Rybchak1, S.Mushasta1

    Author of the translation: N.A.Gorbunov.

    Description: The article describes a hybrid clear neural network that identifies text objects using a parallel incremental scroll block. The neural network then classifies the objects and combines these objects with the original text objects. The method of the proposed neural network is that it can be used both local and global features for processing text data and negotiate with other users. go crazy and increase the number of users compared to a regular network.

    Source: Article: Classification of text tonality using a hybrid convolutional neural network with parallel and serial connections between layers