arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1703.02504cs.CLcs.IRcs.LG
收藏

利用大量弱监督数据进行多语言情感分类

Leveraging Large Amounts of Weakly Supervised Data for Multi-Language Sentiment Classification

ZHAW(苏黎世应用科学大学);ETH Zurich(苏黎世联邦理工学院);Google Research(谷歌研究院);SpinningBytes AG(SpinningBytes公司);EPFL(洛桑联邦理工学院)

Jan Deriu, Aurelien Lucchi, Valeria De Luca, Aliaksei Severyn, Simon Müller, Mark Cieliebak, Thomas Hofmann, Martin Jaggi

摘要:

本文提出了一种新的方法,多语言的情感分类在短文本。这是一项具有挑战性的任务,因为除英语之外的其他语言的训练数据的量非常有限。以前提出的多语言方法通常需要建立一个对应的英语强大的分类器已经可用。相比之下,我们的方法不需要这样的监督。我们利用各种语言的大量弱监督数据来训练多层卷积网络,并展示了使用这种网络的预训练的重要性。我们在各种多语言数据集上彻底评估了我们的方法,包括最近的SemEval-2016情感预测基准(任务4),在那里我们实现了最先进的性能。我们还将为每种语言单独训练的模型的性能与同时为所有语言训练的变体进行了比较。我们发现,后者模型达到略差-但仍然可以接受-性能相比,单一的语言模型,同时受益于更好的泛化性能跨语言。

英文摘要:

This paper presents a novel approach for multi-lingual sentiment classification in short texts. This is a challenging task as the amount of training data in languages other than English is very limited. Previously proposed multi-lingual approaches typically require to establish a correspondence to English for which powerful classifiers are already available. In contrast, our method does not require such supervision. We leverage large amounts of weakly-supervised data in various languages to train a multi-layer convolutional network and demonstrate the importance of using pre-training of such networks. We thoroughly evaluate our approach on various multi-lingual datasets, including the recent SemEval-2016 sentiment prediction benchmark (Task 4), where we achieved state-of-the-art performance. We also compare the performance of our model trained individually for each language to a variant trained for all languages at once. We show that the latter model reaches slightly worse - but still acceptable - performance when compared to the single language model, while benefiting from better generalization properties across languages.

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。

↑