In this work, we develop and evaluate a wide range of feature spaces for deriving Levinstyle verb classifications (Levin, 1993). We perform the classification experiments using Bayesian Multinomial Regression (an efficient log-linear modeling framework which we found to outperform SVMs for this task) with the proposed feature spaces. Our experiments suggest that subcategorization frames are not the most effective features for automatic verb classification. A mixture of syntactic information and lexical information works best for this task. .
