Keyphrase extraction from the Russian corpus on linguistics by means of KEA and RAKE algorithms

P. 369–372.

Moskvina Anna, Sokolova E., Mitrofanova O.

This paper is devoted to comparison of two state-of-the-art keyphrase extraction algorithms, namely KEA based on machine learning and RAKE working with morphosyntactic patterns. Comparative study deal with peculiarities of KEA and RAKE with regard to particular research tasks. Experiments carried out on the Russian corpus on Linguistics allow to work out the best options for testing KEA and RAKE and to find out which algorithm shows better results.

Keywords: keyword extraction, RAKE, KEA, keyphrase

In book

Data Analytics and Management in Data Intensive Domains. Proceedings of the XX International Conference – DAMDID/RCDL’2018, October 9-12, 2018, Moscow

Dudarev V.A., Kiselyova N. N., Temkin I. O. M.: FRC CSC RAS, 2018.