FUSION
FUnctionality Sharing In Open eNvironments
Heinz Nixdorf Chair for Distributed Information Systems
 

Domain specific Information and Relation Extraction: A Hybrid Approach for Entity Recognition in The Ecological Domain

Title: Domain specific Information and Relation Extraction: A Hybrid Approach for Entity Recognition in The Ecological Domain
Author(s): Vladimir Udovenko
Supervisor(s): Alsayed Algergawy and Birgitta König-Ries
School: Friedrich-Schiller-Universität Jena
Thesis Type: Master
Publication Place: Jena
Date: 2019-01-17
Abstract: A prerequisite for managing and integrating scientific data is the ability to identify and determine the same named entity represented in different data sources. Given that scientific data are heterogeneous and represented in different formats, the majority of them are text. Therefore, identifying and extracting entities as well as relations among them plays a crucial role in managing scientific data. To this end, we introduce a hybrid (property-based) approach to extract entities and their relations from scientific publications. The preliminary results demonstrate the effectiveness of the proposed approach.