The 'Path' to Clarity: Identifying False Claims Through a Knowledge Graph Exploration

Automated fact-checking has emerged as a safeguard against the spread of false information. Existing fact-checking approaches aim to determine whether a news claim is true or false, and they have achieved decent accuracy of veracity prediction. However, the current state-of-the-art models still face challenges, such as ambiguity in the claims and lack of contextual information. This study introduces a fact-checking model, Path-FC, which focuses on 1) augmenting the representations of claims and evidence by incorporating additional context using the Knowledge Paths extracted from the external Knowledge Graph; 2) Identifying false claims by learning the differences between claims and evidence. The experimental results demonstrate that Knowledge Path retrieval, combined with the multi-head attention technique, contributes to improved performance of fact-checking. The code is available at https://anonymous.4open.science/r/Path-FC.