On the issue of combining anaphoricity determination and antecedent identification in anaphora resolution
On the issue of combining anaphoricity determination and antecedent identification in anaphora resolution
复制标题
论照应消解中照应度判定与先行词识别相结合的问题
DOI：
10.1109/nlpke.2005.1598742
复制
发表时间：
2005
期刊：
影响因子：
--
通讯作者：
Y. Matsumoto
中科院分区：
文献类型：
--
作者：
R. Iida;Kentaro Inui;Y. Matsumoto
我们提出了一种基于机器学习的名词短语回指解析方法，该方法结合了以前基于学习的模型的优点，同时克服了它们的缺点。我们的回指解析过程与Ng和Cardie提出的分类搜索模型中的步骤顺序相反，但继承了该模型的所有优点。我们进行了解决日语名词短语回指的实验。结果表明，通过基于分类和搜索的修改，我们提出的模型优于早期基于学习的方法。
We propose a machine learning-based approach to noun phrase anaphora resolution that combines the advantages of previous learning-based models while overcoming their drawbacks. Our anaphora resolution process reverses the order of the steps in the classification-and-search model proposed by Ng and Cardie, but inherits all the advantages of that model. We conducted experiments on resolving noun phrase anaphora in Japanese. The results show that with the classification-and-search based modifications, our proposed model outperforms earlier learning-based approaches.