Design and development of an assisted communication system, using machine learning and natural language processing, for the detection of fake news on Twitter.

Abstract

Misleading news has always existed in the world, it has become much easier to spread false information in new communication technologies, allowing users to be not only consumers, but also producers of discourses that circulate and that are often of dubious source of information, known as "Fake News" and these have proliferated around the world. Twitter, a social network that allows users to read and share information, is not free from this problem and, as it is used as an information tool in Latin American media, the spread of fake news is the order of the day, which is why the use of APIs provides a unique value by allowing the communication of various modules, improving the efficiency of use and reliability of the information, The objective of this research is the creation of an extension that allows the detection of false news at a Hispanic level through its own datasets and machine learning, in turn allowing communication with an order bot that reports the results obtained to the staff in charge, allowing to com ply with an efficient flow of assistance.