INFORMATION RELIABILITY TAXONOMY

A CONCEPTUAL PROPOSAL BASED ON THE ANALYSIS OF FACT-CHECKING PLATFORMS

Abstract

The expansion of digital information flows and the circulation of disinformation have intensified the challenges associated with information evaluation in digital environments. Fact-checking platforms address this issue through different procedures of investigation, analysis, and validation, employing classificatory tools to synthesize verification results and guide public interpretation of the analyzed content. This article investigates these semantic schemes in order to understand their patterns and underlying principles. The study adopts an exploratory and qualitative approach, combining documentary research and content analysis of 13 platforms identified through the Duke Reporters’ Lab directory.The results reveal recurring classificatory patterns related to factual accuracy, evidence, verifiability, and informational manipulation, as well as different logical-semantic relations among categories, which, beyond the predominance of hierarchical structures, also include opposition, gradation, and specialization. The analysis further identifies structural limitations, such as terminological heterogeneity, semantic overlap between categories, and the absence of explicit definitions regarding the dimensions and principles applied. A tendency toward classificatory expansion was also observed in more recent platforms, which have incorporated categories associated with multimodal manipulation, content decontextualization, and synthetic forms of disinformation, such as deepfakes. The study defines these systems as information reliability taxonomies, understood as classificatory structures designed to organize informational contents and objects according to evaluative rather than thematic criteria. The findings contribute to broadening discussions on emerging classificatory systems and on the role of Knowledge Organization in mediating information reliability within digital environments.