Fake News Detection Using Machine Learning

• Published in: 2023 International Conference on Artificial Intelligence and Smart Communication (AISC) pp. 759 - 762
• Main Authors: Rawat, Garima, Pandey, Tejaswika, Singh, Tarushi, Yadav, Sonal, Aggarwal, Punnet Kumar
• Format: Conference Proceeding
• Language: English
• Published: IEEE 27.01.2023
• Subjects: Cyber Security; Data models; Feature extraction; Machine learning; Machine learning algorithms; Scripts; Social networking (online); Testing; Uniform resource locators; Voting
• DOI: 10.1109/AISC56616.2023.10085488

In the previous ten years, the volume of information shared online, particularly on social networks, has increased tremendously. The phenomenon of fake news has grown to be a serious issue that jeopardizes the legitimacy of these social networks. The use of machine learning (ML) techniques offers a potential remedy for this issue. To this end, Recently, a number of approaches and algorithms that employ machine learning to identify fake news produced by various platforms supporting social media have been put out in the literature. In order to assess and compile research on the use of machine learning techniques to identify fake news, this chapter will undertake a comprehensive mapping analysis. False propaganda on social media and other platforms is widespread which is a cause of great worry because it can wreak widespread social and national damage with devastating effects. On figuring things, there has already been a lot of research. In order to model a product using supervised machine learning algorithms that can categories fake news as genuine or false using the necessary tools, a survey of the literature on fake news detection is presented in the article. Classic machine learning models are also explored. Feature extraction and vectorization are the results of this technique. We advise using this package to tokenize and extract functions from text input in Python because it has useful tools like the count vectorizer and tiff vectorizer. Then, using feature selection approaches, we investigate and select the most appropriate features to obtain the best accuracy based on the Confusion Matrix results.