Fact-Checking Vietnamese Information Using Knowledge Graph, Datalog, and KG-BERT

In the era of digital information, ensuring accuracy and reliability information is crucial, making fact-checking a vital process. Currently, English has thrived due to various language processing tools ample datasets. However, same cannot be said for Vietnamese fact-checking, which faces significant challenges lack such resources. To address these challenges, we propose model checking facts by synthesizing three popular technologies: Knowledge Graph (KG), Datalog, KG-BERT. The KG serves as foundation process, containing dataset information. logical programming language, used with inference rules complete knowledge within KG. KG-BERT, Deep Learning (DL) model, then trained on this rapidly accurately classify that needs fact-checking. Furthermore, put complex sentences into present solution extracting triples from sentences. This approach also contributes significantly ease constructing foundational datasets evaluate model's performance, create comprising 130,190 samples populate Using enrich graph additional knowledge. utilized train KG-BERT achieving an impressive 95%. Our proposed shows great promise potential contribute development techniques other languages. Overall, research makes contribution field data science providing accurate in contexts.