Svitlana Volkova, a data scientist at Pacific Northwest National Laboratory, is part of a team of researchers who analyzed cryptocurrency discussions on Reddit. (PNNL Photo)

Disarming Disinformation by Understanding Reddit Discussions

It seems the very platforms that we use to share information may be the way we can correct the disinformation or misinformation that floods our timelines.

In an article found at geekwire.com, Alan Boyle writes that: Computer scientists from the Pacific Northwest National Laboratory have mapped the ebb and flow of Reddit’s discussions about cryptocurrency — not only to see how online chatter can predict market behavior but also to gain insights into how disinformation goes viral.

“Cryptocurrency is a very good proxy program for disinformation,” said PNNL data scientist Svitlana Volkova, one of the authors of a study presented at the Web Conference 2019 in San Francisco.

The ups and downs of cryptocurrencies have been much in the news over the past couple of years. The disinformation campaigns orchestrated by Russian agents during the 2016 presidential campaign, however, attracted serious attention to the issue. Even now, cybersecurity experts say the disinformation battle is ramping up for 2020.

It is not a stretch to say everything and nothing is believed on social media platforms most of the time. It is difficult to be sure what is the truth and what is a falsehood anymore. From politics to celebrities being presented in CGI laden porn videos to the latest way to make a buck from home. What is real? And how do you verify what you believe to be real?

When it comes to cryptocurrency, it involves huge amounts of money that attract huge amounts of hackers and thieves. Having the right information for investing is critical.

The PNNL team, Volkova, Emily Saldanha and Maria Glenski — conducted an analysis of tens of thousands of Reddit comments made on the forums for three leading crypto coins between 2015 and 2018. The results were interesting and perhaps not expected.

Some coins got the most comments. Some coins had the longest threads. Some coins had little to no movement.

“These social signals are quite useful, and by incorporating them with machine and deep learning, we intend to build predictive models that hit on causal relationships between different variables so we can explain model decision-making processes,” Volkova said in a news release.

Boyle wrapped up his article with: Network science is making significant headway in understanding how social networks are structured, and how those structures influence the flow of information. But the data will take time to parse to defeat disinformation.

read more at geekwire.com