A statistical clustering software for multi-dimensional clusters

PAC - a statistical clustering software for multi-dimensional clusters

Helmut Schmid

PAC is a predicate-argument clustering software that is trained on predicate-frame-argument tuples and outputs a multi-dimensional cluster analysis, including clusters for the predicates and selectional preference abstraction over the predicate arguments.

PAC is trained on tuples such as

2 abandon SUBJ:NP company project
1 abstain SUBJ legislator
1 accede SUBJ:P:NP government to pact

and induces a statistical soft clustering of these (and other) tuples.


Sabine Schulte im Walde, Christian Hying, Christian Scheible and Helmut Schmid: Combining EM Training and the MDL Principle for an Automatic Verb Classification Incorporating Selectional Preferences. In: Proceedings of ACL-HLT 2008. Columbus, Ohio.


The PAC pages are maintained by Helmut Schmid.


Kontakt IMS

Pfaffenwaldring 5 b, 70569 Stuttgart


Webmaster des IMS

Zum Seitenanfang