First Steps in Czech Entity Linking
Michal Konkol
Text, Speech, and Dialogue (2015)
PDF
Abstract
In this paper, we present our approach for a simplified Entity Linking task in Czech, where entity mentions found in text are linked to a list of known entities. We evaluate both known and newly proposed methods for entity names similarity on a manually annotated newspaper corpus. We show that it is possible to achieve a very high accuracy in this task, which is required in many natural language processing tasks as well as in the commercial practice.
