Paper: Automatic Rule Acquisition for Chinese Intra-chunk Relations

ACL ID I08-2079
Title Automatic Rule Acquisition for Chinese Intra-chunk Relations
Venue International Joint Conference on Natural Language Processing
Session Main Conference
Year 2008
Authors

Multiword chunking is defined as a task to automatically analyze the external function and internal structure of the multiword chunk(MWC) in a sentence. To deal with this problem, we proposed a rule acquisition algorithm to automatically learn a chunk rule base, under the support of a large scale annotated corpus and a lexical knowledge base. We also proposed an expectation precision index to objectively evaluate the descriptive capabilities of the refined rule base. Some experimental results indicate that the algorithm can acquire about 9% useful expanded rules to cover 86% annotated positive examples, and improve the expectation precision from 51% to 83%. These rules can be used to build an efficient rule-based Chinese MWC parser.