DiscoverText

  • Home
  • About
    • Tutorials
    • Customers
    • Testimonials
    • Text Analytics
    • Publications DiscoverText
    • Publications CAT
    • Terms of Service
    • Privacy Policy
    • French Explainer Video
    • Arabic Explainer Video
    • Spanish Explainer Video
  • Products
    • Solutions
    • Features
    • Support
    • NodeXL
  • Industries
    • Consumer
    • Education
    • Government
    • Human Resources
    • Legal
    • Medical & Pharma
  • Blog
  • Free Trial
  • Support
  • Demo
  • Contact Us
  • Login

Tips for Students Working on Fake News Detection Projects

February 24, 2017 by Stuart Shulman 7 Comments

Tweet

We now have about 150 students working in small teams to build fake news detection models. From the field reports, they are having lots of fun learning new ways to think about news, facts, data, tools, and model building. Some of the oservations in this report by George McIntire may be useful for people thinking about the challenge. One paragraph that caught my eye:

“One big assumption underlying this project is that there is considerable overlap in the topics covered by each class of article. If certain words or phrases show up more often in “real” news instead of “fake” news it doesn’t necessarily mean that those terms are associated more with “real” news, but instead could just mean that those words are used in topics more common in the real news dataset. For example, the names of politicians showed up more often in articles classified as “real” than in ones classified as “fake”. I would be committing a huge error if I came to the conclusion that articles mentioning politicians are significantly more likely to be to factual. Some of the most prominent and outlandish fake news articles circulating throughout the internet are ludicrous conspiracy theories about politicians.”

Filed Under: Uncategorized

Comments

  1. android says

    March 29, 2017 at 12:02 pm

    First of all let me tell you, you have got a great blog. I am interested in looking for more of such topics and would like to have further information. Hope to see the next blog soon.

    Reply
  2. android says

    April 11, 2017 at 10:34 am

    We are a group of volunteers and starting a new initiative in a community. Your blog provided us valuable information to work on. You have done a marvelous job!

    Reply
  3. nanda barthare says

    July 6, 2018 at 8:43 pm

    i am also intrested for this type of project.please guide me

    Reply
    • Stuart Shulman says

      August 11, 2018 at 2:40 pm

      Please send us an email at: help@discovertext.com

      Reply
    • Stuart Shulman says

      December 12, 2018 at 10:39 am

      Please write directly to Stu@texifter.com for more information.

      Reply
  4. Nikunj Agarwal says

    August 20, 2018 at 9:40 am

    Can you please provide us with the dataset for fake news detection.

    Reply
    • Stuart Shulman says

      December 12, 2018 at 10:38 am

      Please write to stu@texifter.com about this request. We do have sets we can share.

      Reply

Leave a Reply to Stuart Shulman Cancel reply

Your email address will not be published. Required fields are marked *

Start a Free Trial Request a Web Demo

Featured Posts

Bots, Trolls, and Elections

Tweet

Oh, Canada? A quick research note: your election is definitely under attack via Twitter, despite assertions to the contrary. This is part one of a nine-part series written in the run-up to the 2019 Canadian election. The research was reported by CTV and CBC. Twitter Data Collection on Canadian Elections For the last month, we […]

108 Authenticated Reviews

Tweet

We have been blessed with an engaging group of software users who wrote 108 authenticated DiscoverText reviews. Find out why researchers across the disciplines trust the platform.

Academic Citations

Tweet

We have long worked with diverse and talented scholars to support their publication efforts with free open source and affordable commercial software with academic discounts, free trainings, and almost unlimited enthusiasm for the challenge of capturing and coding text data. We are very pleased to announce a new crop of DiscoverText citations. You can see […]

Recent Posts

  • Top 100 Most Viral Tulsi OR Gabbard Tweets
  • Trump Trolls and Tulsi2020
  • Tulsi Gabbard and MAGA on Twitter
  • Oh Canada: Before You Vote…
  • Bots, Trolls and Elections: Part Nine

Archives

  • November 2019
  • October 2019
  • September 2019
  • August 2019
  • July 2019
  • April 2019
  • March 2019
  • February 2019
  • January 2019
  • December 2018
  • November 2018
  • October 2018
  • August 2018
  • July 2018
  • May 2018
  • April 2018
  • March 2018
  • January 2018
  • December 2017
  • October 2017
  • April 2017
  • March 2017
  • February 2017
  • January 2017
  • December 2016
  • September 2016
  • June 2016
  • January 2016
  • December 2015
  • November 2015
  • October 2015
  • August 2015
  • May 2015
  • April 2015
  • February 2015
  • January 2015
  • December 2014
  • November 2014
  • October 2014
  • September 2014
  • August 2014
  • July 2014
  • June 2014
  • May 2014
  • April 2014
  • March 2014
  • February 2014
  • January 2014
  • December 2013
  • October 2013
  • April 2013
  • December 2012
  • October 2012
  • August 2012
  • July 2012
  • June 2012
  • May 2012
  • April 2012
  • March 2012
  • February 2012
  • January 2012
  • December 2011
  • November 2011
  • October 2011
  • September 2011
  • August 2011
  • May 2011

Boolean defined search, n-grams, word clouds, and custom topic dictionaries

Power tools for text analysis and machine-learning

Discover central topics and also elusive but valuable unexpected or rare concepts. Use this information to train machine-learning classifiers to recognize relevant text and social media data. Jump into data using an interactive word CloudExplorer or build a mini topic dictionary using “defined” search. Try our new CloudExplorer listview for seeing the top 300 bigrams or trigrams in your data. Merge or delete terms as part of building your custom text analytics model.

DiscoverText is powered by uClassify

Machine-learning for everyone

Create gold standard training sets by labeling your training data accurately and reliably using our state-of-the-art collaborative annotation system. Then use our trusted, multilingual machine learning web service (uClassify) to create and apply your own custom-trained text classifiers. Please take the time to check out the work being done by the large and growing uClassify community.

Copyright © 2019 · DiscoverText · 237 Shutesbury Rd.
Amherst, MA 01002 · Powered by ThriveHive