Investigating Content-based Fake News Detection using Knowledge Graphs

Detta är en Master-uppsats från Göteborgs universitet/Institutionen för data- och informationsteknik

Sammanfattning: In recent years, fake news has become a pervasive reality of global news consumption.While research on fake news detection is ongoing, smaller languages such as Swedishare often left exposed by an under-representation in research. The biggest challengelies in detecting news that is continuously shape-shifting to look just like the realthing — powered by increasingly complex generative algorithms such as GPT-2.Fact-checking may have a much larger role to play in the future. To that end,this project considers knowledge graph embedding models that are trained on newsarticles from the 2016 U.S. Presidential Elections. In this project, we show thatincomplete knowledge graphs created from only a small set of news articles candetect fake news with an F-score of 0.74 for previously seen entities and relations.We also show that the model trained on English language data provides some usefulinsights for labelling Swedish-language news articles of the same event domain andsame time horizon.

  HÄR KAN DU HÄMTA UPPSATSEN I FULLTEXT. (följ länken till nästa sida)