arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.01563cs.CL
收藏

基于BERT的跨句子上下文命名实体识别

Exploring Cross-sentence Contexts for Named Entity Recognition with BERT

University of Turku(图尔库大学)

Jouni Luoma, Sampo Pyysalo

摘要:

命名实体识别(NER)通常被视为序列分类任务,其中每个输入由一个文本句子组成。然而,很明显,对任务有用的信息往往可以在单句上下文的范围之外找到。最近提出的自注意模型,如BERT,既可以有效地捕捉输入中的长距离关系,也可以表示由多个句子组成的输入,为在自然语言处理任务中引入跨句子信息的方法创造了新的机会。在本文中,我们提出了一个系统的研究,探索使用跨句信息的NER使用BERT模型在五种语言。我们发现,以额外的句子的形式添加上下文到BERT输入系统地增加NER性能的所有测试的语言和模型。在每个输入中包含多个句子还允许我们研究相同句子在不同上下文中的预测。我们提出了一个简单的方法,上下文多数投票(CMV),结合不同的预测句子,并证明这进一步提高NER性能与BERT。我们的方法不需要对底层BERT架构进行任何更改,而是依赖于重构示例进行训练和预测。对已建立的数据集(包括CoNLL'02和CoNLL'03 NER基准)的评估表明,我们提出的方法可以改善英语,荷兰语和芬兰语的最新NER结果,在德语中实现了最佳的BERT结果,并且与西班牙语中其他基于BERT的方法报告的性能相当。我们在开放许可证下发布本工作中实现的所有方法。

英文摘要:

Named entity recognition (NER) is frequently addressed as a sequence classification task where each input consists of one sentence of text. It is nevertheless clear that useful information for the task can often be found outside of the scope of a single-sentence context. Recently proposed self-attention models such as BERT can both efficiently capture long-distance relationships in input as well as represent inputs consisting of several sentences, creating new opportunitites for approaches that incorporate cross-sentence information in natural language processing tasks. In this paper, we present a systematic study exploring the use of cross-sentence information for NER using BERT models in five languages. We find that adding context in the form of additional sentences to BERT input systematically increases NER performance on all of the tested languages and models. Including multiple sentences in each input also allows us to study the predictions of the same sentences in different contexts. We propose a straightforward method, Contextual Majority Voting (CMV), to combine different predictions for sentences and demonstrate this to further increase NER performance with BERT. Our approach does not require any changes to the underlying BERT architecture, rather relying on restructuring examples for training and prediction. Evaluation on established datasets, including the CoNLL'02 and CoNLL'03 NER benchmarks, demonstrates that our proposed approach can improve on the state-of-the-art NER results on English, Dutch, and Finnish, achieves the best reported BERT-based results on German, and is on par with performance reported with other BERT-based approaches in Spanish. We release all methods implemented in this work under open licenses.

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。

↑