Комбинирование признаКов для извлечения однословных терминов Нокель М. А. (mnokel@gmail.com), МГУ, Москва, Россия Большакова Е. И. (eibolshakova@gmail.com), МГУ, Москва, Россия Лукашевич Н. В. (louk_nat@mail.ru), НИВЦ МГУ, Москва, Россия Ключевые слова: однословные термины, извлечение терминов, ком- бинирование признаков, машинное обучение COMBINING MULTIPLE FEATURES FOR SINGLE-WORD TERM EXTRACTION Nokel M. A. (mnokel@gmail.com) Moscow State University, Moscow, Russia Bolshakova E. I. (eibolshakova@gmail.com) Moscow State University, Moscow, Russia Loukachevitch N. V. (louk_nat@mail.ru) Research Computing Center, Moscow State University, Moscow, Russia The paper describes experiments on automatic single-word term extraction based on combining various features of words, mainly linguistic and statisti- cal, by machine learning methods. Since single-word terms are much more difcult to recognize than multi-word terms, a broad range of word features was taken into account, among them are widely-known measures (such as TF-IDF), some novel features, as well as proposed modifcations of fea- tures usually applied for multi-word term extraction. A large target collection of Russian texts in the domain of banking was taken for experiments. Average Precision was chosen to evaluate the results of term extraction, along with the manually created thesaurus of terminol- ogy on banking activity that was used to approve extracted terms. The experiments showed that the use of multiple features signifcantly im- proves the results of automatic extraction of domain-specifc terms. It was proved that logistic regression is the best machine learning method for sin- gle-word term extraction; the subset of word features signifcant for term extraction was also revealed. Key words: single-word, term extraction, combining features, machine learning