ABSTRACT
Fake news is predominantly influencing people’s perceptions or knowledge to distort their awareness and decision-making. The growth of social media has prevailed fake news causing it to easily blend with truthful information. This project aims a novel text analytics driven approach for detecting fake news and reduce the spread of misinformation. We collect legitimate and fake news, which is transformed from a document based corpus into a topic and event–based representation. Fake news detection is performed using a two-layered approach, which is comprised of detecting fake topics and fake events. By timely detection of misinformation we can control wide dissemination of false or inaccurate content.
Fake news are made by creating non-existent news or adjusting true blue news. The validity of fake news are boosted by mirroring well known authors’ composing styles or communicating suppositions with a tone as often as possible utilised in genuine news. Exceptionally as of late, an expanding number of fake news discovery strategies have been created. All existing location plans may be gathered into two unmistakable classes, to be specific, etymological based strategies and arrange based strategies. Network based approaches for fake news discovery apply arrange properties as a supporting component for different phonetic based approaches. Commonly utilised arrange properties incorporate, but not constrained to site data, authors/subscribers data, time stamps.

PROPOSED SYSTEM
The implementation of the proposed approach for identifying fake news is carried out in two particular stages to find tricky news. To begin with, reliable news are categorised into clusters agreeing to points. Each cluster is centred around common news subjects. Moment, we distinguish fake news by confirming occasions extricated from the news in a particular cluster.
This approach treats news as a fake one in case it may be a news exception (i.e., not classified in any subject cluster) or the likeness between the news occasions and those of the cluster is underneath indicated edge. A expansive number of confirmed news articles classified into news clusters based on subjects and put away in a news database that intermittently gets news upgrades by amassing most recent news stories from genuine news sources such as CNN and Fox News that have been confirmed as authentic by the investigate community. On the off chance that an approaching news to be identified cannot be classified into any existing news cluster, it is checked as a candidate fake news. Something else, the approaching news is put into the comparing cluster for encourage examination. The validity of the approaching news is measured by comparing the occasions extricated from the news with those within the news cluster. When the news article’s validity is underneath a indicated edge, the news are classified as fake.

