Computational fact checking from knowledge networks

TL;DR

This paper presents a scalable computational method for fact checking by analyzing shortest paths in knowledge graphs, which can effectively distinguish true claims from false ones across various domains.

Contribution

It introduces a network-based approach using semantic proximity metrics on knowledge graphs for automated fact checking, demonstrating its effectiveness on Wikipedia data.

Findings

True statements show higher support scores than false ones.

The method is efficient and scalable for large datasets.

It advances automated fact checking to help combat misinformation.

Abstract

Traditional fact checking by expert journalists cannot keep up with the enormous volume of information that is now generated online. Computational fact checking may significantly enhance our ability to evaluate the veracity of dubious information. Here we show that the complexities of human fact checking can be approximated quite well by finding the shortest path between concept nodes under properly defined semantic proximity metrics on knowledge graphs. Framed as a network problem this approach is feasible with efficient computational techniques. We evaluate this approach by examining tens of thousands of claims related to history, entertainment, geography, and biographical information using a public knowledge graph extracted from Wikipedia. Statements independently known to be true consistently receive higher support via our method than do false ones. These findings represent a significant step toward scalable computational fact-checking methods that may one day mitigate the spread of harmful misinformation.