- Oct 2021
-
dl.acm.org dl.acm.org
-
We propose a tri-relationship embedding framework TriFN, which models publisher-news relations and user-news interactions simultaneously for fake news classification. We conduct experiments on two real-world datasets, which demonstrate that the proposed approach significantly outperforms other baseline methods for fake news detection.
It was said in the conclusion that the TriFN can have a good fake news detection performance in the early stage of information dissemination because of the interactions in social media. User credibility was also mentioned since low credibility users tend to spread fake news.
This means that users play a big part in detecting and reducing fake news in social media. Let's be responsible to only share credible news articles and report the misleading ones.
-
-
ieeexplore.ieee.org ieeexplore.ieee.org
-
The rationale is that stories that mix true and false facts may represent attempts to mislead readers. Thus, we focus our analysis on understanding how features can be used to discriminate true and fake news.
I think the mix of true and false information is the most difficult to detect since the false information is intentionally in-between facts and articles on social media feed on surface reading to cause misleading. Even basic google searching can be tricked this way because the algorithm will most likely show results related the keywords on the facts, and not proving that an information is false.
-
FAKE NEWS DETECTION IN PRACTICE
The article showed the scientific processes that can be used in analyzing information and how they applied it in fact-checking. Technology makes fact-checking easier and faster but humans are still the most accurate. That is why studying information science is important because of its relevance to the society.
-