arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.08620cs.CLcs.AIcs.CYcs.HCcs.LG
收藏

POTATO:便携式文本注释工具

POTATO: The Portable Text Annotation Tool

University of Michigan(密歇根大学);University of California, Berkeley(加州大学伯克利分校);The American College of Greece(美国希腊学院)

Jiaxin Pei, Aparna Ananthasubramaniam, Xingyao Wang, Naitian Zhou, Jackson Sargent, Apostolos Dedeloudis, David Jurgens

摘要:

我们提出POTATO,便携式文本注释工具,一个免费的,完全开源的注释系统,1)支持标记多种类型的文本和多模态数据; 2)提供易于配置的功能,以最大限度地提高部署者和注释者的生产力(常见ML/NLP任务、主动学习、按键快捷方式、关键字突出显示、工具提示的方便模板);以及3)支持高度定制(可编辑的UI、插入预筛选问题、注意力和资格测试)。两个标注任务的实验表明,POTATO通过其专门设计的生产力功能提高了标注速度,特别是对于长文档和复杂任务。POTATO可在https://github.com/davidjurgens/potato上获得,并将继续更新。

英文摘要:

We present POTATO, the Portable text annotation tool, a free, fully open-sourced annotation system that 1) supports labeling many types of text and multimodal data; 2) offers easy-to-configure features to maximize the productivity of both deployers and annotators (convenient templates for common ML/NLP tasks, active learning, keypress shortcuts, keyword highlights, tooltips); and 3) supports a high degree of customization (editable UI, inserting pre-screening questions, attention and qualification tests). Experiments over two annotation tasks suggest that POTATO improves labeling speed through its specially-designed productivity features, especially for long documents and complex tasks. POTATO is available at https://github.com/davidjurgens/potato and will continue to be updated.

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。

↑