A pilot study on comparing and extracting impact relations

0Citations
Citations of this article
3Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Documents often contain knowledge about who did what to whom under what conditions, which can be expressed as relations between two entities. Impact relations between two entities express the impacts of one entity on the other. The goal of this pilot study is to examine whether impact relations are similar across domains or not, and investigate how to extract impact relations from unstructured documents using existing techniques. Impact relations in three domains – medical science, international relations, and environmental science (particularly oil spill) are collected and examined. Impact relations account for a significant percentage of semantic relations in all the three domains. The three domains share a common set of impact relations, and each two of the three domains share a significant number of common impact relations. An approach to applying the knowledge to extract impact relations from environmental science documents is proposed. The common impact relations and synonyms of two very different domains can be applied to extract impact relations of a third domain.

Cite

CITATION STYLE

APA

Wu, Y., & Yang, L. (2017). A pilot study on comparing and extracting impact relations. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10647 LNCS, pp. 142–149). Springer Verlag. https://doi.org/10.1007/978-3-319-70232-2_12

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free