Skip to content

Analyzing and clustering tweets to detect disinformation. Includes data preprocessing, topic modeling using BERTopic, and experimentation with multiple models.

Notifications You must be signed in to change notification settings

SERGI0HERREROS/Progress-Task-3-ML

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

36 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Progress Task ML - 3

Repository for the practices of the subject Machine Learning

🤗 Authors

Name Github user
Sergio Herreros Fernández @SergioHerreros
Francisco Javier Luna Ortiz @Lunao01
Carlos Romero Navarro @KarManiatic
Tatsiana Shelepen @Naschkatzee

🎯 Problem description

Disinformation is a hazardous and far-reaching phenomenon, capable of causing profound changes in any community’s political, economic, and cultural framework and thus undermining the foundations of societies around the world, either intentionally or through unconscious mistakes. There is no consensus for a standard definition of this phenomenon involving untruthful information. In the Anglo-Saxon world, the most acknowledged classification is the one proposed in Wardle & Derakhshan in 2017, which divides the alteration or manipulation of information into three typologies:

  • Misinformation: information that is false or misleading but not intended to cause harm.
  • Disinformation: malicious false information, i.e., with a motivation to cause harm.
  • Malinformation: information that is truthful but disseminated with the aim of causing damage.

🚩 Labels

There are two target variables:

  • real - Whether the tweet is real
  • fake - Whether the tweet is fale

It is not going to be relevant when applying BERTopic (labels not consdiered in clustering (unsupervised learning).

🗒️ Features

  • The dataset contains as feature the tweets scraped.

About

Analyzing and clustering tweets to detect disinformation. Includes data preprocessing, topic modeling using BERTopic, and experimentation with multiple models.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •