arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2210.17440cs.CL
收藏

包含命名实体的事实文本中的语义新颖性检测与刻画

Semantic Novelty Detection and Characterization in Factual Text Involving Named Entities

University of Illinois at Chicago(伊利诺伊大学芝加哥分校);Intel Labs(英特尔实验室);PAR Government Systems Corporation(PAR政府系统公司)

Nianzu Ma, Sahisnu Mazumder, Alexander Politowicz, Bing Liu, Eric Robertson, Scott Grigsby

摘要:

现有的文本新颖性检测的大部分工作都是在主题级别上进行研究的,即,识别文档或句子的主题是否新颖。在细粒度语义级别(或上下文级别)上做的工作很少。例如,我们知道埃隆·马斯克是一家科技公司的首席执行官,“埃隆·马斯克在情景喜剧《生活大爆炸》中扮演角色”这句话是新颖和令人惊讶的,因为通常情况下,首席执行官不会是演员。现有的基于主题的新颖性检测方法在这个问题上效果不佳,因为它们不执行涉及文本中的命名实体与其背景知识之间的关系的语义推理。本文提出了一个有效的模型(称为PAT-SND)来解决这个问题,它也可以表征的新颖性。还创建了带注释的数据集。评价表明,PAT-SND的表现大大超过10个基线。

英文摘要:

Much of the existing work on text novelty detection has been studied at the topic level, i.e., identifying whether the topic of a document or a sentence is novel or not. Little work has been done at the fine-grained semantic level (or contextual level). For example, given that we know Elon Musk is the CEO of a technology company, the sentence "Elon Musk acted in the sitcom The Big Bang Theory" is novel and surprising because normally a CEO would not be an actor. Existing topic-based novelty detection methods work poorly on this problem because they do not perform semantic reasoning involving relations between named entities in the text and their background knowledge. This paper proposes an effective model (called PAT-SND) to solve the problem, which can also characterize the novelty. An annotated dataset is also created. Evaluation shows that PAT-SND outperforms 10 baselines by large margins.

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。

↑