UDC 004.02
EFFICIENCY COMPARISON OF METHODS TO COLLECT AND USE NATURAL LANGUAGE MODELS IN MEDICINE
N. G. Vorobyov, post graduate student, Moscow Polytechnic University, Moscow, Russia;
orcid.org/0000-0001-5995-7526, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
The aim of this study is to evaluate methods for autonomous collection and usage of local contexts when selecting a response option by intelligent system during dialog interaction. The analysis uses texts from technical documentation as well as scientific publications related to medi cal subject area. Modern methods to collect local contexts, as well as methods to formalize text data for sub sequent processing by software are considered. A new method for graph modeling of logical relationships in natural language data is proposed. This formalization preserves logical connections between elements and surrounding text making it useful both for linguistic research aimed at studying the specifics of word usage in various fields and for optimizing re sponse selection algorithms in dialogue systems. The efficiency of various approaches to modeling local con texts has been compared.
Key words: : Artificial Intelligence, Local context, Dialog interaction, Thesaurus, Chatbot.