Abstract:
The problem of fake news is a big issue nowadays, with difficulties being caused by false
information. This study examines the use of machine learning (ML) to detect fake
information. Fake news, which often appears real, makes people believe things that aren't
true and causes problems in politics, society, and health. Regular fact-checking struggles
to keep up with the quick spread of online misinformation, The traditional method of factchecking faces limitations in keeping pace with the rapid dissemination of misinformation
online, highlighting the need for more advanced and efficient approaches. So, using
Machine Learning (ML) seems like a better way to deal with this problem. In this study,
we focus on making different ML programs better at finding fake news. These programs
include the 98.9% accurate Bi-Directional Long Short-Term Memory (Bi-Directional
LSTM) and the 99.2% accurate LSTM with Word Embedding Model, Gated Recurrent
Unit (GRU) Model with 98% accuracy, and Recurrent Neural Network (RNN) with an
accuracy of 99.03%, are chosen because they are good at understanding the order of words,
catching language details, and figuring out the context—important for telling if news is
real or fake, make it little it long from the last sentence.
