Skip to content

Fake News! Detector aims to find a way to weed out the real news from the fake news. We create this machine learning model to identify authentic news from the swam of fake news on social media.

Notifications You must be signed in to change notification settings

aliciasply/Final-Project

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

“Fake News!” Detector

Try out our Fake News Detector!

Logo

Background and Motivation:

In the age of social media, it has become more important than ever to be wary of what we read on all platforms. Social media has made it easier to connect with the world and has great influence on the masses as anyone can share anything. What we wanted to do was to find a way to weed out the real news from the fake news. And, more importantly, we wanted to find a way to measure the impact that fake news has on society as a whole. In doing so, we created a machine learning model that predicts if articles are "real" and "fake". With the model created, we ran through thousands of articles posted on Facebook from top news sources (BBC, Buzzfeed Politics, Conservative Post, CNN, Daily Mail, and Fox News). Read our analysis here!

Questions to Answer:

  • Which of these news sites contribute the most “Fake" news?
  • Which of these news sites contribute to the most "Real" news?
  • What are the top 5 most liked "Real"/"Fake" news articles?
  • What are the top 5 most commented "Real"/"Fake" news articles?
  • What are the top 5 most shared "Real"/"Fake" news articles?
  • What do the most engaged "Real"/"Fake" news articles have in common?
  • What are the most-used words in "Fake" news Vs. "Real" news?

Technology

  • Machine Learning: sklearn (train_test_split, TfidfVectorizer, PassiveAggressiveClassifier, accuracy score, confusion_matrix, Pipeline, MultinomialNB)
  • Python/Pandas/Numpy
  • Flask (request/jsonify)
  • CORS
  • Newspaper (Article)
  • urllib
  • NLTK (punkt)
  • HTML/CSS/Bootstrap
  • Octoparse (webscraping tool)
  • Heroku

Outline

  • Create machine learning model
  • Transform, Train & Test the Data
  • Analyze findings from Facebook Data
  • Visualizations
  • Website Creation
  • Create Flask app
  • Deployment to Heroku

Data

Resources

Data Analytics Team:

  • Miguel - Data collection via webscraping, Data cleaning, Flask creation and Heroku Deployment
  • Paola - Data manipulation, Machine learning model creation, Flask creation, Logo Creator
  • Alicia - Data cleaning, Data manipulation, Website creation and Visualizations
  • Medha - Data cleaning, Data manipulation, Visualizations

About

Fake News! Detector aims to find a way to weed out the real news from the fake news. We create this machine learning model to identify authentic news from the swam of fake news on social media.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •