OKOleg Kolobov

Oleg Kolobov

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Oleg Kolobov has a total of 12 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • The information system to control current scientific activities at the Federal Research Center of Information and Computational Technologies

    by Federal Research Center of Information and Computational Technologies, Oleg S. Kolobov, Yulia V. Leonova, Elena V. Rychkova, Olga A. Fedotova, Igor Y. Turchanovsky

    The authors examine the issues of further development of the information system of current scientific activities control at the Federal Research Center of Information and Computational Technologies, and suggest the upgraded model for scientific publications based on several basic principles. They discuss in detail the approach to information retrieval and extraction based on the abstract model of attributes and expressive intuitive query language. The in-house machine-readable format enables to solve the applied problems, e. g. data export/import, data transformation into standard formats, and report generation.

  • Подход к построению рекомендательной системы для библиотек

    by Ю.в. Леонова, О.с. Колобов, А.а. Князева, И.ю. Турчановский

    В работе рассмотрена возможность построения гибридной рекомендательной системы для электронного каталога библиотеки в форме рекомендательного сервиса. Были рассмотрены следующие варианты построения рекомендаций: методы коллаборативной фильтрации, рекомендации на основе контента. Полученные результаты использовались при создании рекомендательного сервиса, в котором предлагается задействовать два рекомендательных алгоритма: коллаборативную фильтрацию на основе документов и рекомендации на основе контента. В качестве входных данных для рекомендательной системы используются анонимизированные данные о выполненных заказах (история заказов) и данные электронного каталога библиотеки. Рассмотрен пример построения рекомендательной системы на основе контента. The paper considers the possibility of building a hybrid recommendation system for the electronic library catalog in the form of a recommendation service. The following options for building recommendations were considered: collaborative filtering methods, content-based recommendations. The results obtained were used to create a recommendation service, in which it is proposed to use two recommendation algorithms: collaborative filtering based on documents and recommendations based on content. Anonymized data on completed orders (order history) and data from the library's electronic catalog are used as input data for the recommendation system. An example of building a recommendation system based on content is considered.

  • An example of empirical approach for bibliographic record linkage

    by Anna Knyazeva, Oleg Kolobov, Igor Turchanovsky

    The record linkage problem in application to a bibliographic and authority data is considered. The problem is common in the situation of merging data from several libraries. The two approaches based on empirical analysis of data are tested. Both of them involve an indirect information about a person. The proposed variant of the decision tree method allows us to deal with inconsistent bibliographic data and to use particular rules one by one for improving of record linkage quality. The study was performed on data of several Russian libraries. The data we deal with are in RUSMARC format which is a variant of UNIMARC popular in Russia.


  • Open Access

    An Example of Automatic Authority Control

    by Anna Knyazeva, Oleg Kolobov, Igor Turchanovsky

    The automatic authority control problem is considered. One possible solution is to use the record linkage approach for authority and bibliographic records. The main aim of this paper is to figure out which concepts and methods are most useful for dealing with our data. An approach based on machine learning method (classification) is considered. A comparative study of different distances and feature sets is made. A study carried out on data of several Russian libraries. The data we deal with are in RUSMARC format which is a variant of UNIMARC popular in Russia.


  • Open Access

    An Instrument for Merging of Bibliographic Databases

    by Anna A. Knyazeva, Oleg S. Kolobov, Fjodor E. Tatarsky, Igor Yu. Turchanovsky

    The process of merging two or more library catalogues is considered in this paper. It's necessary to solve the problem of duplicate detection and merging into one database instead of simple union of different resources. The toolbox Cflib for duplicate detection and merging has been developed by us. It's based on standard principles of record linkage and has quite simple architecture.


  • A merging approach for authority records

    by Anna Knyazeva, Oleg Kolobov, Fjodor Tatarsky, Igor Turchanovsky

    The process of merging two or more authority databases is considered in this paper. We study the problem of duplicate detection and merging authority records, created in accordance with Russian Cataloguing Rules and format RUSMARC/RUSMARC Authorities. The toolbox Cflib has been developed by us for solving this problem. It's based on standard principles of record linkage and has quite simple architecture.