Abstract:

Nowadays, many different tools can be used to classify images, each of which is aimed at solving a certain range of tasks. This article provides a brief overview of libraries and technologies for image classification. The architecture of a simple convolutional neural network for image classification is built. Image recognition experiments have been conducted with popular neural networks such as VGG 16 and ResNet 50. Both neural networks have shown good results. However, ResNet 50 overfitted due to the fact that the dataset contained the same type of images for training, since this neural network has more layers that allow reading the attributes of objects in the images. A comparative analysis of image recognition specially prepared for this experiment was carried out with the trained models.

Abstract:

Methods for creating digital mathematical collections that include unstructured sets of documents are presented. These sets contain materials from scientific conferences, as well as articles from the archives of mathematical journals of the "pre-digital" period.

Using the software tools of the metadata factory of the digital mathematical library Lobachevskii DML, a mandatory set of metadata for digital collection documents was formed. To refine and replenish the metadata sets, knowledge extraction methods from Wikidata were used.

To search Wikidata for information about digital collection documents and their authors, a system of SPARQL queries has been developed. A set of Wikidata entities is defined, which determine the features of the search, as well as the subsequent filtering of the results.

Methods for clarifying and supplementing the bibliographic references given in the articles are proposed. When forming the metadata of documents of retrocollections, a search was made in Wikidata for information about the years of life of the authors of articles, as well as URLs of web pages with information about articles and their authors. The results of the formation of several new digital collections of the Lobachevskii-DML digital library are presented.

Abstract: This article is devoted to some aspects of development a system for managing of graphic documents collections. We describe the administrate and user interfaces of worked system, existing and new image features and their usage in different methods for search and classification, faced problems and solutions to solve them.

Abstract: В данной статье на примере имеющегося у автора опыта развития электронных информационных систем для научно-образовательного сообщества иллюстрируются признаки формирования нового феномена: открытой электронной библиотеки, появление которого является результатом естественной эволюции и системного развития процессов создания, хранения и использования электронных информационных ресурсов.
Автор выражает искреннюю благодарность Фонду Форда, Российскому Фонду Гуманитарных Исследований, Программе интеграционных проектов СО РАН за поддержку исследований и разработок, начиная с 1997 г., а также всем участникам данных работ.