Acquiring Strongly-related Events using Predicate-argument Co-occurring Statistics and Case Frames
Tomohide Shibata, Sadao Kurohashi · International Joint Conference on Natural Language Processing · 2011
This paper proposes a method for automatically acquiring strongly-related events from a large corpus using predicateargument co-occurring statistics and case frames. The strongly-related events are acquired in the form of strongly-related two predicates with their relevant arguments. First, strongly-related events are acquired from predicate-argument cooccurring statistics. Then, the remaining argument alignment is performed by using case frames. We conducted experiments using a Web corpus consisting of 1.6G sentences. The accuracy for the extracted event pairs was 96%, and the accuracy of the argument alignment was 79%. The number of acquired event pairs was about 20 thousands.