Papers by Olga Tikhobaeva

Copied to clipboard

Challenge:	Existing datasets with annotations of scientific terms and relations are difficult to find for other fields, such as biomedical and multi-domains.
Approach:	They present a dataset that includes annotations for two tasks and develop a system called TERMinator for the study of the influence of language models on term recognition.
Outcome:	The proposed system improves the quality of the extracted entities and relations in Russian.