Classification by Compression: Application of Information-Theory Methods for the Identification of Themes of Scientific Texts

I. V. Selivanova; B. Y. A. Ryabko; A. E. Guskov

doi:10.3103/S0005105517030116

Classification by Compression: Application of Information-Theory Methods for the Identification of Themes of Scientific Texts

I. V. Selivanova, B. Y. A. Ryabko, A. E. Guskov

Результат исследования: Научные публикации в периодических изданиях › статья

Аннотация

A method for automatic classification of scientific texts based on data compression is proposed. The method is implemented and investigated based on the data from an archive of scientific texts (arXiv.org) and in the CyberLeninka scientific electronic library (CyberLeninka.ru). Experiments showed that the method correctly identified the themes of scientific texts with a probability of 75-95%; its accuracy depends on the quality of the original data.

Язык оригинала	английский
Страницы (с-по)	120-126
Число страниц	7
Журнал	Automatic documentation and mathematical linguistics
Том	51
Номер выпуска	3
DOI	https://doi.org/10.3103/S0005105517030116
Состояние	Опубликовано - 1 июн 2017

Доступ к документу

10.3103/S0005105517030116

Цитировать

Selivanova, I. V., Ryabko, B. Y. A., & Guskov, A. E. (2017). Classification by Compression: Application of Information-Theory Methods for the Identification of Themes of Scientific Texts. Automatic documentation and mathematical linguistics, 51(3), 120-126. https://doi.org/10.3103/S0005105517030116

@article{f461d05c43f74244a5a727b382e4bc57,

title = "Classification by Compression: Application of Information-Theory Methods for the Identification of Themes of Scientific Texts",

abstract = "A method for automatic classification of scientific texts based on data compression is proposed. The method is implemented and investigated based on the data from an archive of scientific texts (arXiv.org) and in the CyberLeninka scientific electronic library (CyberLeninka.ru). Experiments showed that the method correctly identified the themes of scientific texts with a probability of 75-95%; its accuracy depends on the quality of the original data.",

keywords = "classification, thematic classification of texts, information theory, text compression, arXiv.org, CyberLeninka",

author = "Selivanova, {I. V.} and Ryabko, {B. Y. A.} and Guskov, {A. E.}",

year = "2017",

month = jun,

day = "1",

doi = "10.3103/S0005105517030116",

language = "English",

volume = "51",

pages = "120--126",

journal = "Automatic documentation and mathematical linguistics",

issn = "0005-1055",

publisher = "Allerton Press Inc.",

number = "3",

}

Classification by Compression : Application of Information-Theory Methods for the Identification of Themes of Scientific Texts. / Selivanova, I. V.; Ryabko, B. Y. A.; Guskov, A. E.

В: Automatic documentation and mathematical linguistics, Том 51, № 3, 01.06.2017, стр. 120-126.

Результат исследования: Научные публикации в периодических изданиях › статья

TY - JOUR

T1 - Classification by Compression

T2 - Application of Information-Theory Methods for the Identification of Themes of Scientific Texts

AU - Selivanova, I. V.

AU - Ryabko, B. Y. A.

AU - Guskov, A. E.

PY - 2017/6/1

Y1 - 2017/6/1

N2 - A method for automatic classification of scientific texts based on data compression is proposed. The method is implemented and investigated based on the data from an archive of scientific texts (arXiv.org) and in the CyberLeninka scientific electronic library (CyberLeninka.ru). Experiments showed that the method correctly identified the themes of scientific texts with a probability of 75-95%; its accuracy depends on the quality of the original data.

AB - A method for automatic classification of scientific texts based on data compression is proposed. The method is implemented and investigated based on the data from an archive of scientific texts (arXiv.org) and in the CyberLeninka scientific electronic library (CyberLeninka.ru). Experiments showed that the method correctly identified the themes of scientific texts with a probability of 75-95%; its accuracy depends on the quality of the original data.

KW - classification

KW - thematic classification of texts

KW - information theory

KW - text compression

KW - arXiv.org

KW - CyberLeninka

U2 - 10.3103/S0005105517030116

DO - 10.3103/S0005105517030116

M3 - Article

VL - 51

SP - 120

EP - 126

JO - Automatic documentation and mathematical linguistics

JF - Automatic documentation and mathematical linguistics

SN - 0005-1055

IS - 3

ER -