A Phrasematic Key to the Russian Verb-Nominal Constructions: The Workshop of the Computational Phraseography

The article focuses on the problem of the phrasematic status of verb-nominal phrases considered as "potential phrasemes", such as: давить на психику, ходить на работу, забежать на минутку, забраться на крышу, зайти на страницу. The analyzed list includes more than one thousand of such structures. A special approach to their valency autonomy in the context is proposed. We take into account the context as supraphrasematic feature of potential phrasemes. An individual approach to each potential phrasemes in a particular context has a special role to play. This allows creating a special experience that can be converted to computerized linguistic analysis. We propose to consider the possibility of automatization the detection of the phrasematic status of potential phrasemes by using machine learning methods (the artificial intelligence). The described methods allow to automatic retrieval of the potential phrasemes with synchronous recognition of their phrasematic status in particular text and big text corpora of Russian language.The article focuses on the problem of the phrasematic status of verb-nominal phrases considered as "potential phrasemes", such as: давить на психику, ходить на работу, забежать на минутку, забраться на крышу, зайти на страницу. The analyzed list includes more than one thousand of such structures. A special approach to their valency autonomy in the context is proposed. We take into account the context as supraphrasematic feature of potential phrasemes. An individual approach to each potential phrasemes in a particular context has a special role to play. This allows creating a special experience that can be converted to computerized linguistic analysis. We propose to consider the possibility of automatization the detection of the phrasematic status of potential phrasemes by using machine learning methods (the artificial intelligence). The described methods allow to automatic retrieval of the potential phrasemes with synchronous recognition of their phrasematic status in particular text and big text corpora of Russian language.

Keywords:

Phraseme , Collocation , Idiom , Grammeme , Morphosyntax , ۳-Gram , Machine Learning , NLP , Text Corpora.