人工智能培训

搜索

人工智能技术:SGM:多标签分类的序列生成模型(SGM: Sequence Generation Model for Multi-label Classific

[复制链接]
amwufhvk 发表于 2018-6-14 09:11:38 | 显示全部楼层 |阅读模式
amwufhvk 2018-6-14 09:11:38 1995 0 显示全部楼层
人工智能技术:SGM:多标签分类的序列生成模型(SGM: Sequence Generation Model for Multi-label Classification)多标签分类在自然语言处理中是一项重要且具有挑战性的任务。它比标签趋向于相关的单标签分类更为复杂。现有的方法倾向于忽略标签之间的相关性。此外,文本的不同部分可以对预测不同标签有不同的贡献,这不被现有模型考虑。在本文中,我们提出将多标签分类任务看作序列生成问题,并将新的解码器结构应用于序列生成模型来解决它。广泛的实验结果表明我们提出的方法在以前的工作中有很大的优势。实验结果的进一步分析表明,所提出的方法不仅捕获标签之间的相关性,而且在预测不同标签时自动选择最具信息性的单词。
Multi-label classification is an important yet challenging task in naturallanguage processing.It is more complex than single-label classification inthat the labels tend to be correlated.Existing methods tend to ignore thecorrelations between labels.Besides, different parts of the text cancontribute differently for predicting different labels, which is not consideredby existing models.In this paper, we propose to view the multi-labelclassification task as a sequence generation problem, and apply a sequencegeneration model with a novel decoder structure to solve it.Extensiveexperimental results show that our proposed methods outperform previous work bya substantial margin.Further analysis of experimental results demonstratesthat the proposed methods not only capture the correlations between labels, butalso select the most informative words automatically when predicting differentlabels.人工智能技术:SGM:多标签分类的序列生成模型(SGM: Sequence Generation Model for Multi-label Classification) qgRHA70h7eLcgsHG.jpg
URL地址:https://arxiv.org/abs/1806.04822     ----pdf下载地址:https://arxiv.org/pdf/1806.04822    ----人工智能技术:SGM:多标签分类的序列生成模型(SGM: Sequence Generation Model for Multi-label Classification)
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 立即注册

本版积分规则 返回列表 发新帖

amwufhvk当前离线
新手上路

查看:1995 | 回复:0

快速回复 返回顶部 返回列表