CimpleKG: a Continuously Updated Knowledge Graph of Fact-Checks and Related Misinformation


In this work, we introduce CimpleKG as an open and continuously updated semantic resource about previously fact checked claims and examples of associated misinformed posts. It is a continuously updated public knowledge graph that can be used for supporting misinformation research, by linking various previously published static misinformation datasets with daily updated claims verifications from vetted fact-checking organizations and augments them with additional information such as named entities and contextual factors. We define the notion of factors as being textual features that allow a better understanding of documents. We detect these factors using fully trained transformer-based models. We identify emotion, sentiment, political-leaning, covid-related conspiracy theories, and propaganda techniques as factors and integrate them into the CimpleKG.