Abstract: Nowadays, extensive use of social media and the internet has given rise to fake news that is fabricated articles
intentionally build to mislead readers. Generally, Humans are not good at distinguishing facts and fakes that may cause serious damage to government, market, and society. We made our dataset from 5 publicly available da-tasets to avoid bias towards Indian context-based news articles. By using various feature extraction techniques and Machine learning algorithms, we were able to get an accuracy of 93.90%. The proposed method can be used to debunk false information.

Keywords: Natural Language Processing, Fake News Detection, Rumor Detection, Passive-Aggressive Algo-rithm, Countvectorizer, TF-IDF, NLP, Multinomial Naïve Bayes Classifier

| DOI: 10.17148/IARJSET.2022.9123