论文信息 - Acquiring Semantic Classes to Elaborate Attachment Heuristics

Acquiring Semantic Classes to Elaborate Attachment Heuristics

We use an unsupervised method for learning word classes from partially parsed text corpora. A word class consists of those words that can appear in similar contexts of subcategorization. The main objective of this paper will be to describe an evaluation procedure based on attachment resolution. We will evaluate whether the classes that have been previously learnt are useful in that syntactic task.

Pablo Gamallo | Alexandre Agustini | José Gabriel Pereira Lopes

[1] Osamu Yoshie,et al. Web Knowledge Management and Decision Support , 2003, Lecture Notes in Computer Science.

[2] James Pustejovsky,et al. The Generative Lexicon , 1995, CL.

[3] Dekang Lin,et al. Automatic Retrieval and Clustering of Similar Words , 1998, ACL.

[4] Hang Li,et al. Word Clustering and Disambiguation Based on Co-occurrence Data , 1998, COLING.

[5] Céline Rouveirol,et al. Machine Learning: ECML-98 , 1998, Lecture Notes in Computer Science.

[6] Michael Collins,et al. Prepositional Phrase Attachment through a Backed-off Model , 1995, VLC@ACL.

[7] Pablo Gamallo,et al. Selection Restrictions Acquisition for Parsing Improvement , 2001, INAP.

[8] David Faure,et al. ASIUM: Learning subcategorization frames and restrictions of se-18 lection , 1998 .

[9] Patrick Pantel,et al. An Unsupervised Approach to Prepositional Phrase Attachment using Contextually Similar Words , 2000, ACL.

[10] Pablo Gamallo,et al. Learning Subcategorisation Information to Model a Grammar with "Co-restrictions , 2003 .

[11] Adwait Ratnaparkhi,et al. A Maximum Entropy Model for Prepositional Phrase Attachment , 1994, HLT.

[12] Michael Niemann,et al. Determining PP attachment through semantic associations and preferences , 1998 .

[13] Mats Rooth,et al. Structural Ambiguity and Lexical Relations , 1991, ACL.