中文摘要：

在自然语言中，动词的用法主要包含3种：字面、转喻和隐喻．动词的隐喻用法在隐喻的研究中占有重要的地位．动词的隐喻计算包含两个子任务：隐喻识别和隐喻理解．本文采用条件随机场、最大熵模型来识别动词的隐喻搭配，并在此基础上，加入了两种语义信息：《同义词词林》的同义词信息和《知网》的语义信息来识别动词的隐喻表达，实验识别正确率可达94．47％．

英文摘要：

In a natural language, the expressions of verbs contain three types ： literal, metonymic and metaphoric. Verb metaphor plays an important role in the field of metaphor study. The computing of verb metaphor can he divided into two subtasks： metaphor recognition and metaphor interpretation. This paper uses two models： Conditional Random Fields model and Maximum Entropy model to recognize verb metaphor. And in these basic models, this paper adds two kinds of semantic knowledge ： synonymy information of Tongyici cilin and semantic information of HowNet to recognize verb metaphor. The accuracy of experiment reaches 94.47%.