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 2009.07203cs.DBcs.LG
收藏

CorDEL:一种用于实体链接的对比深度学习方法

CorDEL: A Contrastive Deep Learning Approach for Entity Linkage

Texas A&M University(德州农工大学)

Zhengyang Wang, Bunyamin Sisman, Hao Wei, Xin Luna Dong, Shuiwang Ji

摘要:

实体链接是数据清洗和集成中的一个关键问题。在过去的几十年里,EL通常是由基于规则的系统或具有手工策划功能的传统机器学习模型完成的,这两种模型都严重依赖人工输入。随着新数据的不断增长,已经提出了基于深度学习(DL)的方法来减轻与传统模型相关联的EL的高成本。EL的DL模型的现有探索严格遵循众所周知的双网络架构。然而,我们认为,双网络架构是次优EL,导致现有模型的固有缺陷。为了解决这些缺点,我们提出了一个新颖的和通用的对比DL框架EL。所提出的框架是能够捕捉句法和语义匹配信号,并注意到微妙的,但关键的差异。基于该框架,我们开发了一种用于EL的对比DL方法,称为CorDEL,具有三个强大的变体。我们通过在公共基准数据集和真实数据集上进行的大量实验来评估CorDEL。在公共基准数据集上,CorDEL的性能比之前的最先进模型高出5.2%。此外,CorDEL在真实世界数据集上比当前最佳DL模型提高了2.4%,同时减少了97.6%的训练参数数量。

英文摘要:

Entity linkage (EL) is a critical problem in data cleaning and integration. In the past several decades, EL has typically been done by rule-based systems or traditional machine learning models with hand-curated features, both of which heavily depend on manual human inputs. With the ever-increasing growth of new data, deep learning (DL) based approaches have been proposed to alleviate the high cost of EL associated with the traditional models. Existing exploration of DL models for EL strictly follows the well-known twin-network architecture. However, we argue that the twin-network architecture is sub-optimal to EL, leading to inherent drawbacks of existing models. In order to address the drawbacks, we propose a novel and generic contrastive DL framework for EL. The proposed framework is able to capture both syntactic and semantic matching signals and pays attention to subtle but critical differences. Based on the framework, we develop a contrastive DL approach for EL, called CorDEL, with three powerful variants. We evaluate CorDEL with extensive experiments conducted on both public benchmark datasets and a real-world dataset. CorDEL outperforms previous state-of-the-art models by 5.2% on public benchmark datasets. Moreover, CorDEL yields a 2.4% improvement over the current best DL model on the real-world dataset, while reducing the number of training parameters by 97.6%.

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。

↑