arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1606.01341cs.CL
收藏

用于细粒度实体类型分类的神经结构

Neural Architectures for Fine-grained Entity Type Classification

Tohoku University(东北大学);University College London(伦敦大学学院)

Sonse Shimaoka, Pontus Stenetorp, Kentaro Inui, Sebastian Riedel

摘要:

在这项工作中,我们研究了几个神经网络架构的细粒度实体类型分类。特别是,我们认为最近提出的专注的神经架构的扩展,并作出三个关键的贡献。先前关于注意神经架构的工作没有考虑手工制作的功能,我们将学习和手工制作的功能结合起来,并观察到它们相互补充。此外,通过定量分析,我们建立了注意力机制是能够学习参加超过句法头和短语包含提到,其中都是已知的强大的手工制作的功能,为我们的任务。我们通过分层标签编码方法实现参数共享,该方法在低维投影中为每个类型层次结构显示清晰的聚类。最后,尽管使用相同的评估数据集,但文献经常比较使用不同数据训练的模型。我们建立了训练数据的选择对性能有着巨大的影响,对于先前提出的方法,降低了多达9.85%的松散微F1分数。尽管如此,我们的最佳模型在完善的FIGER(GOLD)数据集上获得了75.36%松散微观F1评分的最新结果。

英文摘要:

In this work, we investigate several neural network architectures for fine-grained entity type classification. Particularly, we consider extensions to a recently proposed attentive neural architecture and make three key contributions. Previous work on attentive neural architectures do not consider hand-crafted features, we combine learnt and hand-crafted features and observe that they complement each other. Additionally, through quantitative analysis we establish that the attention mechanism is capable of learning to attend over syntactic heads and the phrase containing the mention, where both are known strong hand-crafted features for our task. We enable parameter sharing through a hierarchical label encoding method, that in low-dimensional projections show clear clusters for each type hierarchy. Lastly, despite using the same evaluation dataset, the literature frequently compare models trained using different data. We establish that the choice of training data has a drastic impact on performance, with decreases by as much as 9.85% loose micro F1 score for a previously proposed method. Despite this, our best model achieves state-of-the-art results with 75.36% loose micro F1 score on the well- established FIGER (GOLD) dataset.

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。

↑