An Effective Assessment of Machine Learning Approaches for Fake News Detection

False news is a growing issue that threatens public confidence in government institutions and impacts daily lives. Disseminating misleading information is frequently occurring, as newspapers often manipulate information to serve their own agenda before the internet era. People can now spread incorrect or inaccurate information more easily because of the social media platforms' ease of information sharing. People now find it more challenging to differentiate between reliable and fraudulent sources because fake news accounts and websites are becoming more prevalent. To identify fake news from political information, this study aims to use ML algorithms, specifically Logistic Regression(LR), Naive Bayes(NB) and Random Forest(RF) Algorithm. The algorithms identified several forms of false information and provided accurate news classification. The experimental accuracy from the random forest technique yielded values of 99.06% outperforming Logistic Regression and Naive Bayes algorithms, wherein the Random Forest (RF) approach also yielded superior results.