Third-order Variational Reranking on Packed-Shared Dependency Forests
Third-order Variational Reranking on Packed-Shared Dependency Forests
复制标题
DOI：
--
复制
发表时间：
2011-07
期刊：
影响因子：
--
通讯作者：
K. Hayashi;Taro Watanabe;Masayuki Asahara;Yuji Matsumoto
中科院分区：
文献类型：
--
作者：
K. Hayashi;Taro Watanabe;Masayuki Asahara;Yuji Matsumoto
基于Eisner的生成模型，我们提出了一种新的森林重排算法，用于区分依赖句法分析。在我们的框架中，我们定义了两种用于重排序的生成模型。一种是从离线训练数据中学习，另一种是从基线解析器动态生成的森林中学习。在重新排序阶段，使用这些模型的线性内插和判别模型来执行最终预测。为了有效地从表示森林的超图数据结构中训练模型并在其上解码，我们应用了扩展的内部/外部和维特比算法。实验结果表明，与传统方法相比，本文提出的森林重排算法取得了显著的改进。
We propose a novel forest reranking algorithm for discriminative dependency parsing based on a variant of Eisner's generative model. In our framework, we define two kinds of generative model for reranking. One is learned from training data offline and the other from a forest generated by a baseline parser on the fly. The final prediction in the reranking stage is performed using linear interpolation of these models and discriminative model. In order to efficiently train the model from and decode on a hypergraph data structure representing a forest, we apply extended inside/outside and Viterbi algorithms. Experimental results show that our proposed forest reranking algorithm achieves significant improvement when compared with conventional approaches.