ISSN 1991-2927
 

ACP № 2 (60) 2020

Author: "Tatiana Vasilievna Afanaseva"

Tatiana Vasilevna Afanaseva, Ulyanovsk State Technical University, Doctor of Engineering, Associate Professor, Deputy Head of the Department of Information Systems of Ulyanovsk State Technical University; graduated from the Faculty of Radioengineering of UlSTU; an author of articles and monographs in the field of fuzzy modeling and time series analysis. [e-mail: tv.afanasjeva@gmail.com]T. Afanaseva,

Ivan Valerevich Sibirev, State Technical University, Postgraduate Student at the Department of Information Systems of UlSTU; graduated from the Faculty of Economics and Mathematics of UlSTU; an author of articles in the field of fuzzy techniques in multivariate statistical analysis. [e-mail: ivan.sibirev@yandex.ru]I. Sibirev,

Denis Valerevich Zavarzin, Ulyanovsk State Technical University, Postgraduate student at the Department of Information Systems of UlSTU; graduated from the Faculty of Information Systems and Technologies of UlSTU; an author of articles in the field of fuzzy modeling and time series anomaly analysis. [e-mail: dzavarzin@gmail.com]D. Zavarzin

Application of Fuzzy Models in the Analysis of Processes in Organizational-technical Systems 51_11.pdf

The article discusses the application of the Fuzzy models for time series analysis. The main aspect of the analysis is the formation of groups of similar processes. For the grouping of processes, they are considered in the form of discrete time series. Traditional methods for clustering time series based on point-by-point clustering, on attributes or models, do not take into account the nature of the behavior of the time series, which ensures the information content and quality of clustering. To solve this problem, a clustering algorithm based on a fuzzy approach, called Fuzzy Behavior Clustering (FBC approach), was proposed. The proposed time series clustering approach considers the behavior of time series at three levels of the hierarchy (at the level of the general tendency, the level of the parameters and at the level of values). The behavior of time series at the level of the main tendency is described by means of linguistic terms. At the heart of the FBC-approach lie a fuzzy representation of time series and an algorithm for a linguistic assessment of the main tendency. The application of the FBC-approach for analyzing and clustering time series of KPI dynamics of IT project developers based on data from the repository is provided.

Fuzzy behavior clustering, fuzzy clustering behavior, combined clustering of time series, the linguistic terms, extraction of fuzzy tendencies of time series.

2018_ 1

Sections: Computer-aided engineering

Subjects: Computer-aided engineering.


Tatiana Vasilevna Afanaseva, Ulyanovsk State Technical University, Doctor of Engineering; Associate Professor, Deputy Head of Information Systems Department at Ulyanovsk State Technical University; graduated from the Faculty of Radioengineering of Ulyanovsk State Technical University; an author of articles and monographs in the field of the intellectual analysis of time series. [e-mail: tv.afanasjeva@gmail.com]T. Afanaseva,

Aleksei Andreevich Sapunkov, Ulyanovsk State Technical University, Postgraduate Student at the Information Systems Department of Ulyanovsk State Technical University; graduated from the Faculty of Information Systems and Technologies of Ulyanovsk State Technical University; an author of articles in the field of the intellectual analysis of time series. [e-mail: sapalks@gmail.com]A. Sapunkov,

Denis Valerevich Zavarzin, Ulyanovsk State Technical University, Postgraduate Student at the Information Systems Department of Ulyanovsk State Technical University; graduated from the Faculty of Information Systems and Technologies of Ulyanovsk State Technical University; an author of articles in the field of the intellectual analysis of time series. [e-mail: dzavarzin91@gmail.com]D. Zavarzin

Using the K-means Clustering Algorithm for Improving the Temporal Statitics of Commercial Proposals Views 000_6.pdf

Anomalies are considered as not typical and rare values, which decrease accuracy in date significant. Such values would generally cause inaccuracy in date analysis results, so they must be deleted. The article proposes to use the k-means clustering method in order to solve practical problems of data processing for displaying the temporal statistics in the B2B sector. The B2BFamily service for sending and tracking commercial offer represents the subject area and the data source. The article also proposes to remove anomalies and display more adequate temporal view statistics about the average time of the commercial offer slide review. That will help the sales manager to adjust the strategy of communication with customers. Finally, the authors discuss the results and trends of the study further development.

B2bfamily, clustering, anomaly, b2bfamily, k-means clustering algorithm, detection and removal of anomalies.

2016_ 4

Sections: Mathematical modeling

Subjects: Mathematical modeling.


Tatiana Vasilievna Afanaseva, Ulyanovsk State Technical University, Doctor of Engineering; Associate Professor, Deputy Head of Information System Department at Ulyanovsk State Technical University; graduated from the Faculty of Radioengineering at Ulyanovsk State Technical University; an author of articles and monographs in the field of the intellectual analysis of time series. [e-mail: tv.afanasjeva@gmail.com]T. Afanaseva,

Aleksei Andreevich Sapunkov, Ulyanovsk State Technical University, Post-graduate Student of the Information System Department at Ulyanovsk State Technical University; graduated from the Faculty of Information Systems and Technologies at Ulyanovsk State Technical University; an author of articles in the field of the intellectual analysis of time series. [e-mail: sapalks@gmail.com]A. Sapunkov,

Mkrtych Sarkisovich Tonerian, Ulyanovsk State Technical University, Post-graduate Student of the Information System Department at Ulyanovsk State Technical University; graduated from the Faculty of Information Systems and Technologies at Ulyanovsk State Technical University; an author of articles in the field of the intellectual analysis of time series. [e-mail: mkr73@yandex.ru]M. Tonerian

The Two-stage Algorithm of Choosing the Fuzzy Model for Time Series Firecasting 000_8.pdf

The article deals with a two-stage algorithm for the best time-series forecasting model based on the assessment of the model adequacy with the use of behavior and accuracy. For testing, the authors use time series that have been exploited at the International Time Series Forecasting Competition IFAS-EUSLAT in 2015 ([http://irafm.osu.cz/cif/main.php]). Database of this Competition includes 91 numerical time series of different length, tendency, and data reading frequency. Time series values depicted the dynamic of parameters reported from banking area, social networks, and medicine. Three models based on the fuzzy time series concept have been used for time-series forecasting. In order to choose the best model, the two-stage algorithm based on the comparison of time series and model tendencies has been proposed. In addition to the already known quality criterion, the new ones are also exploited in the algorithm. In the conclusion, the results obtained are discussed and the effectiveness of the suggested algorithm is demonstrated.

Fuzzy tendency, fuzzy time series, forecasting, linguistic description.

2015_ 4

Sections: Information systems

Subjects: Information systems.


Nadezhda Glebovna Yarushkina, Ulyanovsk State Technical University, Doctor of Engineering, Professor, First Vice-Rector - Vice-Rector for Science of Ulyanovsk State Technical University; graduated from the Faculty of Radioengineering of Ulyanovsk State Technical University ; an author of more than 250 papers in the field of soft computing, fuzzy logic, and hybrid systems. [e-mail: jng@ulstu.ru]N. Yarushkina,

Tatiana Vasilevna Afanaseva, Ulyanovsk State Technical University, Doctor of Engineering; Associate Professor, Deputy Head of Information Systems Department at Ulyanovsk State Technical University; graduated from the Faculty of Radioengineering at Ulyanovsk State Technical University; an author of articles and monographs in the field of time series data mining. [e-mail: tv.afanasjeva@gmail.com]T. Afanaseva,

Aleksei Mikhailovich Namestnikov, Ulyanovsk State Technical University, Candidate of Engineering, Associate Professor, Associate Professor at the Department of Information Systems at Ulyanovsk State Technical University; graduated from the Faculty of Radioengineering of Ulyanovsk State Technical University; an author of articles and a monograph in the field of intelligent systems for storage and processing of information. [e-mail: nam@ulstu.ru]A. Namestnikov,

Gleb Iurevich Guskov, Ulyanovsk State Technical University, Post-Graduate Student of the Department of Information Systems at Ulyanovsk State Technical University; graduated from the Faculty of Information Systems and Technologies at Ulyanovsk State Technical University; an author of articles in the field of time series data mining. [e-mail: guskovgleb@gmail.com]G. Guskov

Integration of Fuzzy Granular and Ontological Methods for Time Series Analysis 000_8.pdf

The theoretical and methodological foundations of fuzzy granular modeling were developed to solve the problem of fuzzy trend analysis of time series. Their practical implementation as a program complex allows to get the solution of some applied problems. Fuzzy granular presentation of time series includes five levels: from numeric value granulation to the main trend granulation. The authors propose to use fuzzy ontologies of problem area to interpret the results of time series analysis. he problem area fuzzy ontologies were offered to interpret the results of time series analysis. The ontology basis is based on the RDF model that defines classes, instances, ontological relationships, and contingencies. The logical interference of recommendation is realized on the basis of interaction between fuzzy OWL ontology and rule-oriented SWRL rules system. The article deals with the possibilities of integration of several methods of time series forecasting and corresponding aggregates. The integration of techniques for time series analysis and ontological analysis demonstrates the competitiveness of fuzzy trend models based on the mutual reinforcement of ontological and fuzzy granular methods.

Fuzzy time series, fuzzy tendencies, ontology, fuzzy ontology.

2015_ 2

Sections: Information systems

Subjects: Information systems, Artificial intelligence.


Tatyana Vasilievna Afanasieva, Ulyanovsk State Technical University, Doctor of Engineering; graduated from the Radio Engineering Faculty of Ulyanovsk Polytechnic Institute; Professor of Information System Department in Ulyanovsk State Technical University; an author of books and articles in the field of time series data mining and analysis. [e-mail: tv.afanasjeva@gmail.com]T. Afanasieva

Forecasting Time-series Local Trends in the Big Data Analysis 38_5.pdf

According to the forecasts in the field of IT (IDC) the growth of the stored data volumes obtained from various sources will be doubling every two years until 2020. This tendency will remain stable in the conditions of the data increase generated in the OLTP systems, on social networks and by devices in case of mutual exchange of information during the intensive development of the data warehouse, "cloud computing", "the Internet of things" and "digital production" technologies. All that gives rise to a considerable interest in the Big Data analysis and processing from both business and a scientific community. The technology of analytical OLAP-systems for the analysis of the large data, focused on providing a visualization of the multidimensional data and the formation of interactive reports, is one of the most popular in decision support and Business intelligence systems. A promising technology in the field of analysis in addition to the OLAP-systems and aiming at identifying hidden patterns in the large data, is a technology of time series data mining. The most important tasks of time series data mining certainly should include forecasting time series points. A scientific basis of the methodology of the local trends forecasting as fuzzy trends for univariate time series (numeric and non-numeric) which lead to the fuzzy time series are discussed in this article. The solution to this problem is interesting both in theoretical and practical aspects. It is known that fuzzy time series models are proven themselves for a short length time series, and for a long length time series a large computation is required. This paper proposes a k-trend algorithm for extracting local trends out of the big data, considered as a time series, and a time series model in terms of fuzzy local trends is shown. An effect of applying the proposed approach, expressed in a significant reduction of the computational cost when using a fuzzy time series model is shown.

Big data, fuzzy models, time series, local trends, forecasting.

2014_ 4

Sections: Mathematical modeling

Subjects: Mathematical modeling, Automated control systems, Artificial intelligence.


Tatiana Vasilyevna Afanasyeva, Ulyanovsk State Technical University, Candidate of Engineering, Associate Professor at the Chair 'Applied Mathematics and Information Science' of Ulyanovsk State Technical University; author of articles, a monograph, a text-book in the field of intellectual analysis of time series. [e-mail: tv.afanaseva@mail.ru]T. Afanasyeva,

Nadezhda Glebovna Yarushkina, Ulyanovsk State Technical University, Doctor of Engineering, Professor; Pro-Rector for Science; head of the Chair 'Information Systems' at Ulyanovsk State Technical University; author of articles and monographs in the field of intellectual analysis of data. [e-mail: jng@ulstu.ru]N. Yarushkina

Efficiency Analysis of Fuzzy Trend Model for Forecasting of Time Series 26_7.pdf

The article describes a new model for the analysis of time series for the forecast of small time series. The base of the new model is formalization and identification of a new object of time series - a fuzzy trend. The suggested model does not have assumptions used in stochastic simulation, and is easy for the implementation and developed for linguistic interpretation of results. The experiment studies of the accuracy figures of the suggested model reveal its adequacy for forecasting of small time series and competitiveness in comparison with its analogues.

Forecasting, time series, fuzzy trend, accuracy figures.

2011_ 4

Sections: Artificial-intelligence systems

Subjects: Artificial intelligence, Automated control systems, Mathematical modeling.


Nadezhda Glebovna Yarushkina, Ulyanovsk State Technical University, Doctor of Engineering, Professor; Pro-Rector for Science; head of the Chair 'Information Systems' at Ulyanovsk State Technical University; author of articles and monographs in the field of intellectual analysis of data. [e-mail: jng@ulstu.ru]N. Yarushkina,

Valeria Vadimovna Voronina, Ulyanovsk State Technical University, Post-graduate student at the Chair 'Information Systems' of Ulyanovsk State Technical University; graduated from the Faculty of Information Systems and Technology of Ulyanovsk State Technical University; author of articles in the field of intellectual systems for storage and processing of data. [e-mail: vvsh85@mail.ru]V. Voronina,

Tatiana Vasilyevna Afanasyeva, Ulyanovsk State Technical University, Candidate of Engineering, Associate Professor at the Chair 'Applied Mathematics and Information Science' of Ulyanovsk State Technical University; author of articles, a monograph in the field of intellectual analysis of data. [e-mail: tv.afanaseva@mail.ru]T. Afanasyeva

Diagnostics of Helicopter Nodes on Basis of a Model of Grained Time Series 26_8.pdf

In the present paper the authors consider a solution to the problem of diagnostics of helicopter nodes. The diagnostics is carried out by analyzing time series of key physical quantities, based on an expert rulebase containing statements on the significance of trends of change of these variables. In the paper, some expert rules for the helicopter nodes such as a helicopter propulsion engine and main gearbox, are also folmulated.

Diagnostics, time series, helicopters, expert rulebase.

2011_ 4

Sections: Artificial-intelligence systems

Subjects: Artificial intelligence, Information systems.


Tatiana Vasilievna Afanaseva, [e-mail: mars@mv.ru]T. Afanasyeva

Solution of Time-series Data-mining Tasks Within Structural and Linguistic Approach 20_8.pdf

The article gives a description of a new structural and linguistic approach intended for the implementation of time-series data-mining. The orientation of this approach to the analysis of time series of different length and extraction of knowledge on behaviour of time series in the form of fuzzy elementary tendencies as well as the presentation of the results in the linguistic form in bounded natural language, allows extending the group of potential users of systems implemented using the principles of structural and linguistic approach.

Data mining, time series, fuzzy tendency, knowledge extraction, structural and linguis-ticapproach, modeling.

2010_ 2

Sections: Theoretical issues of automation of command and control processes

Subjects: Artificial intelligence, Information systems.


Nadezhda Glebovna Yarushkina, [e-mail: mars@mv.ru]N. Yaroushkina,

Tatiana Vasilievna Afanaseva, [e-mail: mars@mv.ru] T. Afanasyeva,

Irina Grigorievna Perfilyeva, [e-mail: mars@mv.ru] I. Perfilyeva

Integral Method of Fuzzy Modeling and Analysis of Fuzzy Tendencies 20_9.pdf

The article deals with a new method of modeling of time series, which integrates intel-lectual methods of task solution concerning knowledge extraction from time series not only in numerical form but also in the form of linguistic description of levels and elementary tendencies.

Fuzzy model, time series, fuzzy tendency, knowledge extraction, forecast.

2010_ 2

Sections: Theoretical issues of automation of command and control processes

Subjects: Artificial intelligence.


© FRPC JSC 'RPA 'Mars', 2009-2018 The web-site runs on Joomla!