... Jump Right In!
|
| Example: VerbOcean's temporal precedence chains (the "happens-before" relation) between invent and manufacture shown with edge weights |
VerbOcean is a broad-coverage semantic network of verbs.
Here are the the relations the currently released, unrefined version contains, with some data:
| Semantic Relation | Example | Transitive | Symmetric | Num in VerbOcean |
| similarity | produce :: create | Y | Y | 11,515 |
| strength | wound :: kill | Y | N | 4,220 |
| antonymy | open :: close | N | Y | 1,973 |
| enablement | fight :: win | N | N | 393 |
| happens-before | buy :: own; marry :: divorce |
Y | N | 4,205 |
Interactive browser of VerbOcean
For an explanation of how it was constructed, please see:
Timothy Chklovski and Patrick Pantel. 2004.VerbOcean: Mining the Web for Fine-Grained Semantic Verb Relations. In Proceedings of Conference on Empirical Methods in Natural Language Processing (EMNLP-04). Barcelona, Spain. [pdf]
For some work on how to use paths through the graph to refine individual edges, please see:
Timothy Chklovski and Patrick Pantel. 2005. Global Path-based Refinement of Noisy Graphs Applied to Verb Semantics. In Proceedings of The Second International Joint Conference on Natural Language Processing (IJCNLP-05), Jeju Island, South Korea, October 11-13, 2005 [pdf]
Timothy Chklovski and Patrick Pantel. 2004. Path Analysis for Refining Verb Relations. In Proceedings of KDD Workshop on Link Analysis and Group Detection (LinkKDD-04). Seattle, WA. [pdf]