Detección automática de noticias falsas en español con técnicas de machine learning

The work developed in this project stems from the need to somehow stop the growing spread of fake news shared by the general public in digital media. That is why we set out to develop a viable product that helps in the automatic identification of false news with machine learning techniques. For the implementation of this solution it was decided to take into account alternative attributes to directly process the text of the news that could determine the veracity of this, such as the percentage of capital letters in the title or the number of question marks. As a result of the described, a chatbot implemented in Amazon Lex integrated into Facebook Messenger was obtained that receives the title and the text of the news, then these parameters are processed by the machine learning model and finally the user is answered if the news is it true or false and with what probability.

• Autores: Quintero Perozo, Diany Yisel; Ortega Riveros, Juan Alberto
• Tipo de recurso: Trabajo de grado de pregrado
• Fecha de publicación: 2020
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: spa
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/51485
• Acceso en línea: http://hdl.handle.net/1992/51485
• Palabra clave: Noticias falsas; Hábitos de información; Aprendizaje automático (Inteligencia artificial); Ingeniería
• Rights: openAccess
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/