arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1503.00095cs.CL
收藏

面向任务的语义关系分类词嵌入学习

Task-Oriented Learning of Word Embeddings for Semantic Relation Classification

The University of Tokyo(东京大学);University College London(伦敦大学学院);Toyota Technological Institute(丰田工业大学)

Kazuma Hashimoto, Pontus Stenetorp, Makoto Miwa, Yoshimasa Tsuruoka

摘要:

我们提出了一种新的学习方法,词嵌入设计的关系分类。我们的词嵌入是通过在大型未标记语料库上使用词汇关系特定特征预测名词对之间的词来训练的。这使我们能够显式地将特定于关系的信息合并到词嵌入中。然后使用学习的词嵌入来构造用于关系分类模型的特征向量。在一个完善的语义关系分类任务,我们的方法显着优于基线的基础上,以前介绍的词嵌入方法,并比较有利的是以前的国家的最先进的模型,使用语法信息或手动构建的外部资源。

英文摘要:

We present a novel learning method for word embeddings designed for relation classification. Our word embeddings are trained by predicting words between noun pairs using lexical relation-specific features on a large unlabeled corpus. This allows us to explicitly incorporate relation-specific information into the word embeddings. The learned word embeddings are then used to construct feature vectors for a relation classification model. On a well-established semantic relation classification task, our method significantly outperforms a baseline based on a previously introduced word embedding method, and compares favorably to previous state-of-the-art models that use syntactic information or manually constructed external resources.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑