arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1904.13178cs.CL
收藏

具有减少假否定和大类型覆盖的细粒度实体识别

Fine-grained Entity Recognition with Reduced False Negatives and Large Type Coverage

Indian Institute of Technology Guwahati(古瓦哈提印度理工学院);Indira Gandhi Delhi Technical University for Women(英迪拉·甘地德里女子技术大学)

Abhishek Abhishek, Sanya Bathla Taneja, Garima Malik, Ashish Anand, Amit Awekar

摘要:

细粒度实体识别(FgER)是检测和分类实体的任务,涉及到一个大的类型集,跨越不同的领域,如生物医学,金融和体育。我们观察到,当类型集跨越几个域,实体提及的检测成为监督学习模型的限制。主要原因是缺乏适当注释实体边界的数据集,同时覆盖大量实体类型。我们的工作直接解决了这个问题。我们提出了启发式与远程监督(HAnDS)框架自动构建一个质量数据集适合FGER任务。HAnDS框架以流水线方式利用Wikipedia和Freebase之间的高度互连,减少了天真地使用远程监督方法所引入的注释错误。使用HAnDS框架,我们创建了两个数据集,一个适合于构建基于FIGER类型层次结构识别多达118个实体类型的FGER系统,另一个适合于基于TypeNet层次结构识别多达1115个实体类型。我们广泛的实证实验保证了生成的数据集的质量。除此之外,我们还提供了一个手动注释的数据集,用于对FgER系统进行基准测试。

英文摘要:

Fine-grained Entity Recognition (FgER) is the task of detecting and classifying entity mentions to a large set of types spanning diverse domains such as biomedical, finance and sports. We observe that when the type set spans several domains, detection of entity mention becomes a limitation for supervised learning models. The primary reason being lack of dataset where entity boundaries are properly annotated while covering a large spectrum of entity types. Our work directly addresses this issue. We propose Heuristics Allied with Distant Supervision (HAnDS) framework to automatically construct a quality dataset suitable for the FgER task. HAnDS framework exploits the high interlink among Wikipedia and Freebase in a pipelined manner, reducing annotation errors introduced by naively using distant supervision approach. Using HAnDS framework, we create two datasets, one suitable for building FgER systems recognizing up to 118 entity types based on the FIGER type hierarchy and another for up to 1115 entity types based on the TypeNet hierarchy. Our extensive empirical experimentation warrants the quality of the generated datasets. Along with this, we also provide a manually annotated dataset for benchmarking FgER systems.

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。

↑