融合多种方法提升知识图谱事实核查准确率
HybridFC: A Hybrid Fact-Checking Approach for Knowledge Graphs
- 集成文本、路径、规则和嵌入等多种方法
- 在FactBench上AUC提升0.14至0.27
- 适合需要高精度事实核查的系统开发者
我们研究旨在预测知识图谱中断言真实性的事实核查方法。近年来提出了五类主要的事实核查方法，每种都存在部分重叠的局限性。当前基于文本的方法受限于人工特征工程；基于路径和规则的方法仅依赖知识图谱作为背景知识；基于嵌入的方法在现有事实核查任务中准确率较低。为此，我们提出一种混合方法——HybridFC，通过在集成学习框架下利用各类方法的多样性，显著提升预测性能。在FactBench数据集上，该方法相比现有最先进方法，受试者工作特征曲线下面积（AUC）提升了0.14至0.27。代码已开源，地址为https://github.com/dice-group/HybridFC。
原文摘要 · Abstract (English)
We consider fact-checking approaches that aim to predict the veracity of assertions in knowledge graphs. Five main categories of fact-checking approaches for knowledge graphs have been proposed in the recent literature, of which each is subject to partially overlapping limitations. In particular, current text-based approaches are limited by manual feature engineering. Path-based and rule-based approaches are limited by their exclusive use of knowledge graphs as background knowledge, and embedding-based approaches suffer from low accuracy scores on current fact-checking tasks. We propose a hybrid approach -- dubbed HybridFC -- that exploits the diversity of existing categories of fact-checking approaches within an ensemble learning setting to achieve a significantly better prediction performance. In particular, our approach outperforms the state of the art by 0.14 to 0.27 in terms of Area Under the Receiver Operating Characteristic curve on the FactBench dataset. Our code is open-source and can be found at https://github.com/dice-group/HybridFC.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。