KEA (Keyphrase Extraction Algorithm) is one of the well-known algorithms used by researchers working in the field of NLP (Natural Language Processing). KEA is aimed at effective automatic extraction of keywords and phrases. KEA shows good results in processing English texts and proves to be quite useful for linguistic purposes. This algorithm can be very helpful for the users who want to get a general idea of the material they are interested in. We present modification of KEA designed for processing Russian corpora. We linked KEA and necessary linguistic tools for processing Russian texts and then carried out experiments on representative Russian text corpora and examined reliability of our KEA modification. Therefore, in this paper we provide detailed description of KEA’s internal organization, its functional peculiarities, experimental settings, results and evaluation procedure.