Improving Verb Clustering with Automatically Acquired Selectional Preferences
Improving Verb Clustering with Automatically Acquired Selectional Preferences
复制标题
DOI：
10.3115/1699571.1699596
复制
发表时间：
2009-08
期刊：
影响因子：
--
通讯作者：
Lin Sun;A. Korhonen
中科院分区：
文献类型：
--
作者：
Lin Sun;A. Korhonen
在以往的动词自动分类研究中，句法特征被证明是最有用的特征，尽管人工分类在很大程度上依赖于语义特征。我们表明，与以前的工作相比，可以获得相当大的额外的改进，通过使用语义特征的自动分类：动词选择偏好，从语料库数据中获得使用完全无监督的方法。我们报告这些有前途的结果，使用一个新的框架，动词聚类，它结合了最近的次范畴化采集系统，丰富的句法语义特征集，和谱聚类的变化，特别是在高维特征空间。
In previous research in automatic verb classification, syntactic features have proved the most useful features, although manual classifications rely heavily on semantic features. We show, in contrast with previous work, that considerable additional improvement can be obtained by using semantic features in automatic classification: verb selectional preferences acquired from corpus data using a fully unsupervised method. We report these promising results using a new framework for verb clustering which incorporates a recent subcategorization acquisition system, rich syntactic-semantic feature sets, and a variation of spectral clustering which performs particularly well in high dimensional feature space.