Automatic Classification of Semantic Relations between Facts and Opinions

Abstract

Classifying and identifying semantic re-lations between facts and opinions on the Web is of utmost importance for or-ganizing information on the Web, how-ever, this requires consideration of a broader set of semantic relations than are typically handled in Recognizing Tex-tual Entailment (RTE), Cross-document Structure Theory (CST), and similar tasks. In this paper, we describe the con-struction and evaluation of a system that identifies and classifies semantic rela-tions in Internet data. Our system targets a set of semantic relations that have been inspired by CST but that have been gen-eralized and broadened to facilitate ap-plication to mixed fact and opinion data from the Internet. Our system identi-fies these semantic relations in Japanese Web texts using a combination of lexical, syntactic, and semantic information and evaluate our system against gold stan-dard data that was manually constructed for this task. We will release all gold standard data used in training and eval-uation of our system this summer. 1