Detección de noticias falsas en redes sociales basada en aprendizaje automático y profundo: Una breve revisión sistemática

Social networks have changed how society is informed. Social networks such as Twitter and Facebook have millions of users who often share fake news without knowing it. The contents of these news are false and unchecked, and they become viral, deceiving, and causing panic. The objective of this study is to develop a literature review that examines how machine and deep learning have supported the development of social media fake news classifiers. The study was developed from a formal methodology used in computer science. The results showed that learning models have been widely used to create false news detection systems, with detection predominating in the political field. It was found that machine learning models were mostly used in contrast with deep learning models, however, both approaches demonstrated be efficient to classify fake news, playing a decisive factor of the results, the data set and the feature extraction method used.