Abstract | ||
---|---|---|
In this work, we develop and evaluate a wide range of feature spaces for deriving Levin- style verb classifications (Levin, 1993). We perform the classification experiments using Bayesian Multinomial Regression (an effi- cient log-linear modeling framework which we found to outperform SVMs for this task) with the proposed feature spaces. Our exper- iments suggest that subcategorization frames are not the most effective features for auto- matic verb classification. A mixture of syntac- tic information and lexical information works best for this task. |
Year | Venue | Keywords |
---|---|---|
2008 | ACL | feature space,log linear model |
Field | DocType | Volume |
Verb,Subcategorization,Computer science,Multinomial logistic regression,Support vector machine,Artificial intelligence,Natural language processing,Syntax,Machine learning,Bayesian probability | Conference | P08-1 |
Citations | PageRank | References |
11 | 0.54 | 16 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jianguo Li | 1 | 377 | 35.38 |
Chris Brew | 2 | 321 | 44.44 |