A Detecting Fake News Using Machine Learning: A Comparative Study of Techniques

Many people get their news via the Internet and social media platforms, and given the rapid growth of these platforms, fake news may now spread easily and quickly. False information that aims to mislead and harm society and the individual is known as fake news. By deliberately spreading false information, the media distort public opinion and threaten the social order by leading people to believe things that are not true. With the massive expansion of social media networks, the spread of fake news has increased dramatically. Although interesting, it poses some difficulties due to limited resources (such as datasets and published research). This paper presents diverse machine-learning techniques to identify fabricated news by analyzing the textual content. Several techniques were used, including SVM, RF, logistic regression, Naive Bayes, Gradient Boosting, AdaBoost, KNN, DT, and XGBoost. Based on comparing the results, who got the best result with an accuracy rate of 0.9967 and the lowest loss of 0.003 The study includes a variety of methodologies, such as natural language processing (NLP), machine learning, and data mining, which have been found to improve the efficiency of text processing to increase accuracy and can save time and effort by automatically identifying fake news, especially in light of the massive amount from materials available on the Internet.