Detecting Fake News on Social Networks: A Naive Bayes Approach to Enhance Information Trustworthiness

In modern life, the influence of social network takes the first place. The news from a social network may be wrong, in the manner it's the necessary to check the trustworthiness of a news is very important. This article proceeds a novel approach to find a fake news on social media by implementing the machine learning algorithm. The classification of real or fake news using Naïve Bayes classifier and TF-IDF, NLP were used for tokenization, count vectorizer, and so on. Two different datasets (fake & real) were used here. Various types of machine learning algorithms such as logistic regression, SVM, Decision tree and random forest are implemented in the same datasets. While comparing the results Naïve Bayes give 95% of accuracy for the dataset. Various preprocessing techniques were employed here to achieve the results. It is efficient solution to find or classify, a news is real or fake on social media.