March 31, 2000
Extraction and classification of causal relations from semi-technical texts
This work investigates the causal relation as it is expressed in text, more specifically in semi-technical texts. The purpose of semi-technical texts is to explain notions within a particular domain to a non-initiated person, which means they would contain useful information for the domain model. Our goal is twofold, first we aim at gaining some insight into the feasibility of automatic extraction of causal relations from text, and second we make a proposal for the classification of the causal relations to capture the variations expressed in the text.