Nowadays, the term Fake News has stood out all over the world. A good example of this was the Brazilian election in 2018, in which Fake News had a major impact on the election outcome. Considering these facts, it is appropriate to develop a mechanism, by using machine learning techniques, that can classify news as false or true. This paper proposes two methods of classification, convolutional neural networks (CNN) and artificial neural networks of multiple layers (ANN), in the detection of Fake News. These two methods were compared to two others found in the literature, using the same database. While in this work an accuracy of 96% for CNN and 95% for ANN was obtained, in the literature the highest accuracy was 89% for SVM (Support Vector Machine) and 90% for ANN. Resumo: Hoje em dia, o termo Fake News tem se destacado em todo o mundo. Um bom exemplo disso, foi a eleição brasileira de 2018, na qual as Fake News tiveram um grande impacto no resultado das eleições. Considerando esses fatos, torna-se apropriado o desenvolvimento de um mecanismo, por meio das técnicas de aprendizado de máquina, que seja capaz de classificar notícias em falsas ou verdadeiras. Este artigo, propõe dois métodos de classificação, as redes neurais convolucionais (Convolutional Neural Network-CNN) e as redes neurais artificiais de múltiplas camadas (RNA), na detecção das Fake News. Estes dois métodos foram comparados a outros dois encontrados na literatura, utilizando a mesma base de dados. Enquanto neste trabalho obteve-se uma acurácia de 96% para a CNN e 95% para a RNA, na literatura as maiores acurácias foram de 89% para o SVN (Support Vector Machine) e 90% para a RNA.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
334 Leonard St
Brooklyn, NY 11211
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.