Finding Credibility Clues on Twitter

Michael Morguarge

1 Comments

1242

Published By: Georgia Tech News Center, 1/26/2017

>> View the Article <<

Summary

Georgia Tech Researchers have created an early prototype system that uses a language model to predict the perceived credibility of social media messages, for example based on key mood words. This prototype system could lead to an application to help users identify credible information on social media.

Flesch-Kincaid Grade Level of Article: 12.6

Extended Discussion Questions

As a user of social media, how do you determine if a message is credible or not?
In the news release, Eric Gilbert is quoted as saying, “Twitter is part of the problem with spreading untruthful news online.”
- What are some of the ways untruthful news spreads through Twitter and other social media?
- Are there any legal issues that arise from the spreading of untruthful news?
Gilbert then suggests that Twitter “can also be part of the solution” to stopping the flow of untruthful news.
- How could a language model help with filtering false news?
- How does truthful information benefit from social media dissemination?
- Can you think of other ways that Twitter could be part of the solution?
The news release mentions some of the characteristics their system uses to decide whether a social media message should be categorized as credible.
- Are these characteristics enough to judge if messages are credible?
- What other characteristics might you add to improve the language model?
- Besides language, what other factors could be used to rate the credibility of a message?
- Should the system consider the source’s reputation? What kind of data would it need if the source is a news site? What kind of data would it need if the source is an individual user?

Relating This Story to the CSP Curriculum Framework

Global Impact Learning Objectives:

LO 7.1.1 Explain how computing innovations affect communication, interaction, and cognition.
LO 7.3.1 Analyze the beneficial and harmful effects of computing.
LO 7.5.2 Evaluate online and print sources for appropriateness and credibility.

Global Impact Essential Knowledge:

EK 7.1.1C Social media continues to evolve and fosters new ways to communicate.
EK 7.1.1H Social media, such as blogs and Twitter, have enhanced dissemination.
EK 7.1.1N The Internet and the Web have changed many areas, including e-commerce, health care, access to information and entertainment, and online learning.
EK 7.3.1A Innovations enabled by computing raise legal and ethical concerns.
EK 7.5.2A Determining the credibility of a source requires considering and evaluating the reputation and credentials of the author(s), publisher(s), site owner(s), and/or sponsor(s).

Other CSP Big Ideas:

Idea 3 Data and Information

Banner Image: “Network Visualization – Violet – Offset Crop”, derivative work by ICSI. New license: CC BY-SA 4.0. Based on “Social Network Analysis Visualization” by Martin Grandjean. Original license: CC BY-SA 3.0

Tagged: 3 Data & Info, 7.1.1 Interaction and cognition, 7.1.1C Social media, 7.1.1H Social sharing, 7.1.1N Breadth of change, 7.3.1 Benefits and harm, 7.3.1A Law and ethics, 7.5.2 Evaluate sources, 7.5.2A Credibility

This topic has 1 reply, 2 voices, and was last updated 6 years, 1 month ago by Julia Bernd.

Viewing 0 reply threads

Author

Posts
- February 18, 2017 at 2:04 am #2478
  
  Julia Bernd
  Admin
  
  One thing that struck me about this research — at least as it was described in the article — is that the researchers were collapsing a couple different dimensions of credibility in a way that may have affected their results, or at least their interpretation.
  
  For example, they pointed out that lower perceived credibility (by their measure) correlates with more retweeting — but they also pointed out that lower perceived credibility correlates with more hedges. Given that the rating scale is accuracy, people may judge the information contained in a hedged tweet as less likely to be accurate because the tweeter has already introduced uncertainty — but they may be more likely to retweet it, because of a higher perception of the credibility of the source. (Exactly because that source doesn’t seem to be overstating their claims.)
  
  This is only one possible interpretation — and a rather optimistic one at that — but it illustrates an important point: If you’re going to try to do experiments on humans, it’s important to be careful about exactly what question you ask, and to make sure you interpret your results in the light of how that question was worded.
  
  Rather a tangent for a CS class, but it’s a point that computer scientists often miss, especially if they aren’t co-authoring with non-computer scientists. So, this is a plug for interdisciplinary collaboration! 🙂
Author

Posts

Viewing 0 reply threads

You must be logged in to reply to this topic.