AUTOMATIC CONTROL AND COMPUTER SCIENCES最新文献

{"title":"Tasks in Named Entity Recognition: Technologies and Tools","authors":"N. S. Lagutina,&nbsp;A. M. Vasilyev,&nbsp;D. D. Zafievsky","doi":"10.3103/S0146411624700251","DOIUrl":"10.3103/S0146411624700251","url":null,"abstract":"<p>The task of named entity recognition (NER) is to identify and classify words and phrases denoting named entities (NEs), such as people, organizations, geographical names, dates, events, and terms from subject areas. While searching for the best solution, researchers conduct a wide range of experiments with different technologies and input data. A comparison of the results of these experiments shows a significant discrepancy in the quality of NER and poses the problem of determining the conditions and limitations for the application of the used technologies, as well as finding new solutions. An important part in answering these questions is the systematization and analysis of current research and the publication of relevant reviews. In the field of NE recognition, the authors of analytical articles primarily consider mathematical methods of identification and classification and do not pay attention to the specifics of the problem itself. In this survey, the field of NE recognition is considered from the point of view of individual task categories. The authors identify five categories: the classical task of NER, NER subtasks, NER in social media, NER in domain, and NER in natural language processing (NLP) tasks. For each category the authors discuss the quality of the solution, features of the methods, problems, and limitations. Information about current scientific works of each category is given in the form of a table for clarity. This review allows us to draw a number of conclusions. Deep learning methods are the leading methods among state-of-the-art technologies. The main problems are the lack of datasets in open access, strict requirements for computing resources, and the lack of error analysis. A promising area of research in NER is the development of methods based on unsupervised techniques or rule-based learning. Intensively developing language models in existing NLP tools can serve as a possible basis for text preprocessing for NER methods. The article ends with a description and results of experiments with NER tools for Russian-language texts.</p>","PeriodicalId":46238,"journal":{"name":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","volume":"58 7","pages":"779 - 796"},"PeriodicalIF":0.6,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396703","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of Deep Neural Networks for Automatic Irony Detection in Russian-Language Texts","authors":"M. A. Kosterin,&nbsp;I. V. Paramonov","doi":"10.3103/S0146411624700469","DOIUrl":"10.3103/S0146411624700469","url":null,"abstract":"<p>This paper examines automatic methods for classifying Russian-language sentences into two classes: ironic and nonironic. The methods under consideration can be divided into three categories: classifiers based on language model embeddings, classifiers based on sentiment information, and classifiers that train embeddings to detect irony. The components of classifiers are neural networks such as BERT, RoBERTa, BiLSTM, and CNN, as well as an attention mechanism and fully connected layers. Experiments to detect irony are carried out using two corpora of Russian-language sentences: the first corpus is composed of journalistic texts from OpenCorpora, while the second corpus is an extension of the first one and is supplemented with ironic sentences from Wiktionary. The best results are demonstrated by a group of classifiers based on pure embeddings of language models with the maximum F-measure value of 0.84, achieved by a combination of RoBERTa, BiLSTM, an attention mechanism, and a pair of fully connected layers in experiments on an extended corpus. In general, using the extended corpus produces results that are 2–5% better than those using the basic corpus. The achieved results are the best for the problem under consideration for the Russian language and are comparable to the best ones for English.</p>","PeriodicalId":46238,"journal":{"name":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","volume":"58 7","pages":"1073 - 1081"},"PeriodicalIF":0.6,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396472","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LTL-Specification for Development and Verification of Control Programs","authors":"M. V. Neyzov,&nbsp;E. V. Kuzmin","doi":"10.3103/S0146411624700366","DOIUrl":"10.3103/S0146411624700366","url":null,"abstract":"<p>This work continues the series of articles on the development and verification of control programs based on the LTL-specification. The approach consists in describing the behavior of programs by special form of linear temporal logic (LTL) formulae. The developed LTL-specification can be directly verified with the help of a model-checking tool. Next, according to the LTL-specification, a program code is unambiguously built in the imperative programming language. The specification is translated into the program using a template. The novelty of the work is the proposal of two new LTL-specifications, which are declarative and imperative, as well as in a stricter formal justification of this approach to program development and verification. A transition is made to nuXmv, a more advanced verification tool for finite and infinite systems. It is proposed to describe the behavior of control programs in a declarative manner. For this purpose, a declarative LTL-specification is intended, which defines a transition system as a formal model of program behavior. This behavior description method is quite expressive—the theorem on the Turing completeness of the declarative LTL-specification is proved. Next, to build a program code in the imperative language, the declarative LTL-specification is converted to an equivalent imperative LTL-specification. An equivalence theorem is proved, which guarantees that both specifications determine the same behavior. The imperative LTL-specification is translated into an imperative program code according to the presented template. The declarative LTL-specification, which undergoes verification, and the control program based on it are guaranteed to determine the same behavior in the form of a corresponding transition system. Thus, in the verification a model coherent with the actual behavior of the control program is used.</p>","PeriodicalId":46238,"journal":{"name":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","volume":"58 7","pages":"920 - 945"},"PeriodicalIF":0.6,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396499","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Joint Simplification of Various Types of Spatial Objects While Preserving Topological Relations","authors":"O. P. Yakimova,&nbsp;D. M. Murin,&nbsp;V. G. Gorshkov","doi":"10.3103/S0146411624700378","DOIUrl":"10.3103/S0146411624700378","url":null,"abstract":"<p>Cartographic generalization includes the selection of objects and phenomena, as well as their simplification (generalization), to be shown in the map while preserving the main typical traits, characteristics, and correlations by the criteria set by the user in the request. These criteria include the problem being solved and the scale of the displayed map. Various map transformations can change the relationships between objects, especially since it is common practice to simplify each type of spatial object independently (first the administrative boundaries, then the road network, populated localities, hydrographic network, etc.). The solution of topological conflicts is one of the key problems of generalizing digital map. A great deal of attention has been paid to the solution of this problem since the start of studies in this domain. The consideration of coverings and mesh structures confines the more general problem of managing topological conflicts to solve topological conflicts within a mesh cell. This study proposes a new geometric simplification algorithm. Its feature consists of the joint simplification of a set of spatial objects of various types while preserving their topological relations. The only parameter of the proposed algorithm is the minimal size of the detail shown in the map. Usually, this size is equal to one millimeter in the target map scale. Step one of the algorithm is the construction of a special mesh data structure. This structure is used to form, for each spatial object, a sequence of cells, to which points of this object belong. If a cell contains points of only one object, the object is geometrically simplified within the bounding cell using the sleeve-fitting algorithm. If a cell contains points of several objects, the geometric simplification is done using a special topology-preserving procedure.</p>","PeriodicalId":46238,"journal":{"name":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","volume":"58 7","pages":"946 - 956"},"PeriodicalIF":0.6,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Optimized Algorithm of Finding the Shortest Path in a Multiple Graph","authors":"A. V. Smirnov","doi":"10.3103/S0146411624700226","DOIUrl":"10.3103/S0146411624700226","url":null,"abstract":"<p>In the article, we consider undirected multiple graphs of any natural multiplicity <i>k</i> &gt; 1. A multiple graph contains edges of three types: ordinary edges, multiple edges, and multiedges. Each edge of the last two types is a union of <i>k</i> linked edges, which connect 2 or (<i>k</i> + 1) vertices, correspondingly. The linked edges should be used simultaneously. If a vertex is incident to a multiple edge, then it can be incident to other multiple edges, and it can also be the common end of <i>k</i> linked edges of a multiedge. If a vertex is the common end of a multiedge, then it cannot be the common end of another multiedge. As for an ordinary graph, we can define the integer function of the length of an edge for a multiple graph and set the problem of the shortest path joining two vertices. Any multiple path is a union of <i>k</i> ordinary paths adjusted on the linked edges of all multiple and multiedge edges. In this article, the previously obtained algorithm for finding the shortest path in an arbitrary multiple graph is optimized. We show that the optimized algorithm is polynomial. Thus, the shortest path problem is polynomial for any multiple graph.</p>","PeriodicalId":46238,"journal":{"name":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","volume":"58 7","pages":"745 - 752"},"PeriodicalIF":0.6,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Semantic Rule-Based Sentiment Detection Algorithm for Russian Publicism Sentences","authors":"A. Y. Poletaev,&nbsp;I. V. Paramonov,&nbsp;E. I. Boychuk","doi":"10.3103/S0146411624700408","DOIUrl":"10.3103/S0146411624700408","url":null,"abstract":"<p>This article studies the task of sentiment detection in Russian sentences, which is understood as the author’s attitude on the sentence topic expressed through linguistic expression features. Today most studies on this subject utilize texts of a colloquial style, limiting the applicability of their results to other styles of speech, particularly to publicism. To fill the gap, the authors developed new publicism sentences oriented toward a sentiment detection algorithm. The algorithm recursively applies appropriate rules to parts of sentences represented as constituency trees. Most of the rules are proposed by a philologist, based on knowledge of expression features from Russian philology, and are algorithmized using constituency trees generated by the algorithm. A decision tree and sentiment vocabulary are also used in this study. This article contains the results of evaluation of the algorithm on the corpus of publicism sentences OpenSentimentCorpus and the F-measure is 0.80. The results of errors analysis are also presented.</p>","PeriodicalId":46238,"journal":{"name":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","volume":"58 7","pages":"977 - 994"},"PeriodicalIF":0.6,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396502","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Construction of the Optimal Information Exchange Scheme of an Adaptive Motion Control System for a Group of Unmanned Aerial Vehicles","authors":"L. N. Kazakov,&nbsp;E. P. Kubyshkin,&nbsp;D. E. Paley","doi":"10.3103/S0146411624700238","DOIUrl":"10.3103/S0146411624700238","url":null,"abstract":"<p>This paper considers the problem of modeling an adaptive control system information exchange for a group of unmanned aerial vehicles (UAVs). The UAV group moves in accordance with the adaptive algorithm for optimal control. Optimal controls are constructed to ensure that the minimal total energy is expended. The parameters of the mathematical model of the movement of the UAV group are refined during the flight in accordance with the changing external conditions. In accordance with this, the control actions are specified. This task requires significant computing resources and imposes special requirements on the information exchange system between the UAV and the control point. A scheme of the information exchange between the UAV and the control point, which makes it possible to calculate the optimal parameters of the transmitting devices, is proposed.</p>","PeriodicalId":46238,"journal":{"name":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","volume":"58 7","pages":"753 - 761"},"PeriodicalIF":0.6,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Algorithms for the Eulerian Cycle and Eulerian Trail Problems for a Multiple Graph","authors":"A. V. Smirnov","doi":"10.3103/S0146411624700342","DOIUrl":"10.3103/S0146411624700342","url":null,"abstract":"<p>In this article, we consider undirected multiple graphs of any natural multiplicity <i>k</i> &gt; 1. A multiple graph contains edges of three types: ordinary edges, multiple edges, and multiedges. Each edge of the last two types is a union of <i>k</i> linked edges that connect 2 or (<i>k</i> + 1) vertices, correspondingly. The linked edges should be used simultaneously. If a vertex is incident to a multiple edge, then it can be incident to other multiple edges, and it can also be the common end of <i>k</i> linked edges of a multiedge. If a vertex is the common end of a multiedge, then it cannot be the common end of another multiedge. We set the problem of finding the Eulerian walk (the cycle or the trail) in a multiple graph, which generalizes the classical problem for an ordinary graph. The necessary conditions of existence of an Eulerian walk in a multiple graph are formulated; it is shown that these conditions are not sufficient. In addition, it is shown that the necessary conditions of existence of an Eulerian cycle and an Eulerian trail are not mutually exclusive for an arbitrary multiple graph; therefore, it is possible to construct a multiple graph in which two types of Eulerian walks exist simultaneously. Any multiple graph can be juxtaposed to the ordinary graph with quasi-vertices, which represents the structure of the initial graph in a simpler form. In particular, each Eulerian walk in the multiple graph corresponds to the Eulerian walk in the graph with quasi-vertices. The algorithm for constructing such a graph is formulated. The auxiliary problem of finding the covering trails with the given endpoints in an ordinary graph is also considered, and two algorithms for solving it are obtained. We elaborate the algorithm for finding the Eulerian walk in a multiple graph, which has exponential complexity. For the special case of a multiple graph, a polynomial algorithm is proposed; it is shown that in this special case the necessary conditions of existence of an Eulerian walk are sufficient.</p>","PeriodicalId":46238,"journal":{"name":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","volume":"58 7","pages":"889 - 903"},"PeriodicalIF":0.6,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396504","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Annotation of Text Corpora by Sentiment and Irony in a Project of Citizen Science","authors":"I. V. Paramonov,&nbsp;A. Y. Poletaev","doi":"10.3103/S0146411624700263","DOIUrl":"10.3103/S0146411624700263","url":null,"abstract":"<p>This paper studies the construction of a corpus of sentences annotated by general sentiment into four classes (positive, negative, neutral, and mixed), a corpus of phrasemes annotated by sentiment into three classes (positive, negative, and neutral), and a corpus of sentences annotated by the presence or absence of irony. The annotation is conducted by volunteers within the project Preparing Texts for Algorithms on the People of Science website. Based on the available knowledge of the subject area for each of the problems, guidelines for the annotators are compiled. A methodology for the statistical processing of the annotation results is also developed based on analyzing the distributions and agreement measures of the annotations of different annotators. For annotating sentences by irony and phrasemes by sentiment, the agreement measures are quite high (the full agreement rate is 0.60–0.99), while for annotating sentences by general sentiment, the agreement is low (the full agreement rate is 0.40), apparently due to the higher complexity of the problem. It is also shown that the performance of automatic algorithms for sentence sentiment analysis improves by 12–13% when using a corpus on whose sentences all annotators (3–5 people) agree compared with a corpus annotated by only one volunteer.</p>","PeriodicalId":46238,"journal":{"name":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","volume":"58 7","pages":"797 - 807"},"PeriodicalIF":0.6,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396704","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Requirement Patterns in Deductive Verification of poST Programs","authors":"I. M. Chernenko,&nbsp;I. S. Anureev,&nbsp;N. O. Garanina","doi":"10.3103/S0146411624700421","DOIUrl":"10.3103/S0146411624700421","url":null,"abstract":"<p>Process-oriented programming is one of the approaches used to develop control software. A process-oriented program is defined as a sequence of processes. Each process is represented by a set of named states containing a program code that define the logic of the process’ behavior. Program execution is a sequential execution of each of these processes in their current states at each iteration of the control loop. Processes can interact through changing the states of each other and shared variables. This paper develops a method for classifying temporal requirements for process-oriented programs in order to simplify and automate the deductive verification of such programs. The method consists of the following steps. At the first step, the requirements are formalized in a specialized language DV-TRL, a variant of the typed first-order predicate logic with a set of interpreted types and predicate and functional symbols that reflects specific concepts of the control systems in the process-oriented paradigm. At the second step, the formalized requirements are divided into classes, each of which is defined by a pattern—a parametric formula of the DV-TRL language. The verification conditions generated for process-oriented programs with respect to the requirements satisfying the same pattern have the same proof scheme. At the third step, appropriate proof schemes are developed. In our paper, we first give a brief introduction to the poST language, a process-oriented extension to the ST language of the IEC 61131-3 standard. Next, the DV-TRL language is defined. We also provide a collection of natural language requirements for several control systems. Then we define the patterns that fully cover all the requirements of this collection. For each of these patterns, we give an example of a formalized requirement from the collection and describe a scheme for proving the verification conditions for this pattern. St-atistics on the distribution of requirements from the collection over patterns reveals the most popular patterns. We also analyze related works.</p>","PeriodicalId":46238,"journal":{"name":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","volume":"58 7","pages":"1003 - 1024"},"PeriodicalIF":0.6,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396470","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}