Результаты исследований: Научные публикации в периодических изданиях › статья в журнале по материалам конференции › Рецензирование
Keyword extraction from single Russian document. / Sandul, Mikhail Vadimovich; Mikhailova, Elena Georgievna.
в: CEUR Workshop Proceedings, Том 2135, 01.01.2018, стр. 30-36.Результаты исследований: Научные публикации в периодических изданиях › статья в журнале по материалам конференции › Рецензирование
}
TY - JOUR
T1 - Keyword extraction from single Russian document
AU - Sandul, Mikhail Vadimovich
AU - Mikhailova, Elena Georgievna
PY - 2018/1/1
Y1 - 2018/1/1
N2 - The problem of automatic keyword and phrases extraction from a text occurs in different tasks of information retrieval and text mining. The task is the identification of terms that best describe the subject of a document. Currently there are a lot of research to solve this problem. Basically, algorithms are developed for texts in English. The possibility of applying these algorithms to the Russian texts are not sufficiently investigated. One of the most known algorithms for solving the problem of keyword extraction is RAKE. This article examines the effectiveness of RAKE algorithm for texts in Russian. The work also applies the hybrid method, which uses the Γ-index metric for phrases weighting, which were obtained using the algorithm RAKE. The article shows that this algorithm is more accurate than PAKE while reducing the number of selected phrases.
AB - The problem of automatic keyword and phrases extraction from a text occurs in different tasks of information retrieval and text mining. The task is the identification of terms that best describe the subject of a document. Currently there are a lot of research to solve this problem. Basically, algorithms are developed for texts in English. The possibility of applying these algorithms to the Russian texts are not sufficiently investigated. One of the most known algorithms for solving the problem of keyword extraction is RAKE. This article examines the effectiveness of RAKE algorithm for texts in Russian. The work also applies the hybrid method, which uses the Γ-index metric for phrases weighting, which were obtained using the algorithm RAKE. The article shows that this algorithm is more accurate than PAKE while reducing the number of selected phrases.
UR - http://www.scopus.com/inward/record.url?scp=85050482656&partnerID=8YFLogxK
M3 - Conference article
AN - SCOPUS:85050482656
VL - 2135
SP - 30
EP - 36
JO - CEUR Workshop Proceedings
JF - CEUR Workshop Proceedings
SN - 1613-0073
T2 - 3rd Conference on Software Engineering and Information Management, SEIM 2018
Y2 - 14 April 2018
ER -
ID: 38400996