Structural Disambiguation Based on Reliable Estimation of Strength of Association

This paper proposes a new class-based method to estimate the strength of association in word co-occurrence for the purpose of structural disambiguation. To deal with sparseness of data, we use a conceptual dictionary as the source for acquiring upper classes of the words related in the co-occurrence, and then use t-scores to determine a pair of classes to be employed for calculating the strength of association. We have applied our method to determining dependency relations in Japanese and prepositional phrase attachments in English. The experimental results show that the method is sound, effective and useful in resolving structural ambiguities.

• Type: Conference
• Year: 1998
• Where: ACL
• Authors: Haodong Wu, Eduardo de Paiva Alves
• Keywords: ACL 1998, ACL 2007, Prepositional Phrase Attachments, Structural Disambiguation, Word Co-occurrence