Twitter's Information Operations - An OSINT Analysis

Wed, 12 Feb 2020 00:00:00 +0000

Key Takeaways 🔗

Twitter is doing better than other platforms by releasing datasets, albeit partial, on Information Operations (IO).
- There is so much more information yet to be disclosed. Recommendations are given.
Attribution blindspots seem to be a common problem with social media companies.
Aggregated Twitter data and Python scripts are available on Github - and will be kept up-to-date.
Beautiful dynamic data visualization for Twitter’s IO datasets, generated in real time from our GitHub datasets.
A similar study for other platforms such as YouTube would be interesting. Maybe Google’s Threat Analysis Group could start publishing comprehensive datasets? :)

In our last OSINT analysis of Facebook’s Coordinated Inauthentic Behavior we highlighted the pitfalls of Facebook’s data-sharing policies and the lack of transparency when it comes to processes and awareness of influence campaigns on the platform. Although, previous work has been done on some of the Twitter datasets - in this analysis, we extend our work to examine Twitter’s Information Operations (IO) and the measures they are taking (or neglecting) to combat the rampant growth of disinformation, misinformation, and influence campaigns. All the data used in this analysis was downloaded from Twitter’s archives of suspended accounts. The data from Twitter can be accessed on their transparency report, whereas our aggregated data for this analysis is available through GitHub, including the script used to generate the datasets – feel free to send us pull requests.

Information on Matt Suiche

Twitter's Information Operations - An OSINT Analysis

Key Takeaways 🔗