Information laundering works like this:
- Adversaries plant fake news to low quality / less credible media outlets (or anything that produces news)
- They alert more credible news to this news article
- Credible/mainstream news picks up (if one picks up the news, the others generally pick it up as well)
- Profit
The research question is, can we find instances of this scheme on the internet?
We will limit the project to bitcoin/cryptocurrency-related news. You will do the following:
1- Collecting news articles mentioning bitcoin (it’s ok to use social media to do that)
2- Categorizing or ranking these articles’ outlets in terms of credibility or mainstreamness (you can use existing datasets, no need to do ml for that)
3- Determining information flow between news articles/outlets, i.e. compute the textual similarity between articles to determine if they refer to the same news.
4- Does the information flow from less credible to more credible sources? Can we prove information laundering?
Requirements: Basic ML knowledge and data analysis skills in Python.
Please send a C.V. and Transcript!
You do not need to submit anything else beforehand.
Contact: | Tuğrulcan Elmas |