Assisting the Human Fact-Checkers: Detecting All Previously Fact-Checked Claims in a Document

TL;DR

This paper introduces a system to assist human fact-checkers by identifying sentences in documents that match previously verified claims, using a new dataset and learning-to-rank methods to improve accuracy.

Contribution

It presents a novel document-level approach for detecting verifiable claims, along with a new annotated dataset and evaluation metrics, advancing fact-checking automation.

Findings

Learning-to-rank approach improves detection performance

Modeling text similarity and stance enhances accuracy

Veracity considerations are crucial for claim verification

Abstract

Given the recent proliferation of false claims online, there has been a lot of manual fact-checking effort. As this is very time-consuming, human fact-checkers can benefit from tools that can support them and make them more efficient. Here, we focus on building a system that could provide such support. Given an input document, it aims to detect all sentences that contain a claim that can be verified by some previously fact-checked claims (from a given database). The output is a re-ranked list of the document sentences, so that those that can be verified are ranked as high as possible, together with corresponding evidence. Unlike previous work, which has looked into claim retrieval, here we take a document-level perspective. We create a new manually annotated dataset for this task, and we propose suitable evaluation measures. We further experiment with a learning-to-rank approach, achieving sizable performance gains over several strong baselines. Our analysis demonstrates the importance of modeling text similarity and stance, while also taking into account the veracity of the retrieved previously fact-checked claims. We believe that this research would be of interest to fact-checkers, journalists, media, and regulatory authorities.