Mining Answers for Causation Questions

Roxana Girju and Dan Moldovan

Causal relations of various kinds are a pervasive feature of human language and theorising about the world. Despite this, the specification of a satisfactory general analysis of causal relations has long proved difficult. This paper provides a syntactic and semantic classification of cause-effect lexico-syntactic patterns for automatic detection and extraction of causation relationships in English texts. We also present a semiautomatic method of discovering generally applicable lexico-syntactic patterns that refer to the causation relation. The patterns are found automatically, but their validation is done semi-automatically. Our final purpose is to add a new module to our existing Question Answering (QA) system that will answer complex cause-effect questions.

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.