FAKE NEWS DETECTION USING MACHINE LEARNING

Abstract

The increase in the amount of false information is a widespread issue in the era of digital technology, with extensive implications for society such as fostering skepticism, manipulation, and undermining democratic conversations. In order to tackle this pressing matter, this study utilizes the use of NLP strategies and employs an array of ML methods to construct efficient models for detecting false news. The objective is to identify the most effective approach for addressing this issue. The chosen methods, namely logistic regression, naive Bayes, support vector machines, random forests, and k-nearest neighbors, are rigorously evaluated to ascertain their efficacy in identifying counterfeit news. The key findings indicate that the ML techniques are highly effective in differentiating between genuine and fake news stories, achieving accuracy ratings between 85% and 95%. The performance parameters, including precision, recall, and F1-score, are thoroughly examined to offer a full comparison. This research enhances the developing field of false news identification by showcasing the suitability of Natural Language Processing (NLP) and a variety of ML techniques. In addition to academic domains, the study aims to explore practical applications, providing a detailed comprehension of the pros and cons of each algorithm. The study continues by providing insights into potential future paths, highlighting the necessity for flexible strategies in identifying false information, considering the ever-changing nature of disinformation. In summary, our research significantly contributes to the battle against false information by creating efficient detection models and providing useful insights into the capabilities and constraints of various ML methods