Fake News Detection Using Machine Learning

Abstract

It is the purpose of this project to investigate the uses of Natura l
Language Processing (NLP) methods in the identification of
'fake news,' which is misleading news items that originate from
non-reputable sources and are spread on the internet. As
opposed to counting words, you'll need to develop an algo ri thm
that uses a word tallies matrix (which uses word tallies relative to
how often they are used in other articles in your dataset) or a
tfidf matrix (which utilises word tallies relative to how often the y
are used in other articles in your dataset) rather than a count
vectorizer to determine the frequency of use of words in your
dataset. These models, despite the fact that they take into account
crucial qualities such as word ordering and context, fall short in
a number of other ways. It is very possible that two papers with a
comparable word count but vastly different interpretations were
both authored by the same person, according to the rules of
probability. Following the presentation of the problem, members
of the data science community reacted enthusiastically, adopting
proactive actions to remedy the situation as soon as possible.
Among other things, artificial intelligence is helping Facebook
delete fake news items from users' news feeds. The company is
using artificial intelligence as part of a Kaggle competition titled
the "False News Challenge," which is being run by Kaggle.
Fighting fake news is an easy text categorization project wi th a
clear purpose in mind, and it is being carried out as part of a
bigger public education and awareness campaign. Können
Pretend you're in the following situation: In your research, you
have developed a model that can tell the difference between
legitimate news and false or fabricated news. The crea t ion o f a
dataset including both fake and legitimate news items is proposed
for this purpose, following which a Naive Bayes classifier would
be used to discriminate between the two types of news