Investigating Different Syntactic Context Types and Context Representations for Learning Word Embeddings
Investigating Different Syntactic Context Types and Context Representations for Learning Word Embeddings
复制标题
DOI：
10.18653/v1/d17-1257
复制
发表时间：
2017-09
期刊：
影响因子：
--
通讯作者：
Bofang Li;Tao Liu;Zhe Zhao;Buzhou Tang;Aleksandr Drozd;Anna Rogers;Xiaoyong Du
中科院分区：
文献类型：
--
作者：
Bofang Li;Tao Liu;Zhe Zhao;Buzhou Tang;Aleksandr Drozd;Anna Rogers;Xiaoyong Du
单词嵌入模型的数量每年都在增长。它们大多是基于单词及其上下文的共现信息。然而，什么是上下文的最佳定义仍然是一个悬而未决的问题。我们对四种不同的句法语境类型和语境表示进行了系统的研究，以学习词嵌入。通过综合实验来评价其在6项外在和内在任务上的有效性。我们希望本文以及发布的代码能够帮助您为给定的任务选择最佳的上下文类型和表示。
The number of word embedding models is growing every year. Most of them are based on the co-occurrence information of words and their contexts. However, it is still an open question what is the best definition of context. We provide a systematical investigation of 4 different syntactic context types and context representations for learning word embeddings. Comprehensive experiments are conducted to evaluate their effectiveness on 6 extrinsic and intrinsic tasks. We hope that this paper, along with the published code, would be helpful for choosing the best context type and representation for a given task.