KFU. Card publication. Analysis of the cluster structure of collections of mathematical papers with different UDC codes

KAZAN
FEDERAL UNIVERSITY

- RUS
- 中文
- ES

ANALYSIS OF THE CLUSTER STRUCTURE OF COLLECTIONS OF MATHEMATICAL PAPERS WITH DIFFERENT UDC CODES

Form of presentation

Articles in international journals and collections

Year of publication

2023

��

��

Authors, employees of KFU

Nevzorova Olga Avenirovna, author

Authors, students KFU

Gizatullin Bulat Timurovich, author

Bibliographic description in the original language

O. A. Nevzorova, and B. T. Gizatullin Analysis of the cluster structure of collections of mathematical papers with different UDC codes // Lobachevskii Journal of Mathematics, 2022, Vol. 43, No. 12, pp. 3597–3604.

Annotation

Lobachevskii Journal of Mathematics

Keywords

clustering, universal decimal classification, UDC code, mathematical paper

The name of the journal

Lobachevskii Journal of Mathematics

URL

https://link.springer.com/article/10.1134/S1995080222150239#citeas

Please use this ID to quote from or refer to the card

https://repository.kpfu.ru/eng/?p_id=282670&p_lang=2

Full metadata record

Field DC	Value	Language
dc.contributor.author	Nevzorova Olga Avenirovna	ru_RU
dc.contributor.author	Gizatullin Bulat Timurovich	ru_RU
dc.date.accessioned	2023-01-01T00:00:00Z	ru_RU
dc.date.available	2023-01-01T00:00:00Z	ru_RU
dc.date.issued	2023	ru_RU
dc.identifier.citation	O. A. Nevzorova, and B. T. Gizatullin Analysis of the cluster structure of collections of mathematical papers with different UDC codes // Lobachevskii Journal of Mathematics, 2022, Vol. 43, No. 12, pp. 3597–3604.	ru_RU
dc.identifier.uri	https://repository.kpfu.ru/eng/?p_id=282670&p_lang=2	ru_RU
dc.description.abstract	Lobachevskii Journal of Mathematics	ru_RU
dc.description.abstract	Clustering is the task of dividing data objects into groups of similar objects. The influence of the specifics of the texts of scientific articles of one subject area for clustering problems has been little studied at present. This article is devoted to the problem of clustering collections of mathematical papers that have the different Universal Decimal Classification (UDC) codes. The study was carried out on the collection of mathematical papers published in the «Izvestiya VUZov. Matematika” journal for 10 years. The size of this collection is about 1000 original papers with different UDC codes. The collection contains subcollections of papers that have the same UDC code. The objective of our research is to analyze the cluster structure of repre-sentative subcollections of papers with the same UDC code, which will allow us to evaluate various parameters of the constructed clusters in the future. We have performed the standard pre-processing (tokenization, lemmatiza-tion) of Russian math texts. The following text vectorization methods were in-vestigated: tf-idf, doc2vec trained on the original data, and a pretrained word2vec model. All vectors were normalized using the L2 norm. To identify optimal hyperparameters and check the quality of clustering, we have used internal efficiency measures such as Silhouette coefficient, Calinski-Harabaz index. We also used the elbow method for hyperparameters tuning.The following clustering algorithms have been investigated: k-means, agglomerative clustering, affinity propagation, DBSCAN, spectral clustering. The optimal hyperparameters were selected for each method, and then the results of clustering were compared. As a result we have selected the optimal methods of clustering math papers.	ru_RU
dc.language.iso	ru	ru_RU
dc.subject	clustering	ru_RU
dc.subject	universal decimal classification	ru_RU
dc.subject	UDC code	ru_RU
dc.subject	mathematical paper	ru_RU
dc.title	Analysis of the cluster structure of collections of mathematical papers with different UDC codes	ru_RU
dc.type	Articles in international journals and collections	ru_RU