By Michael Lieberman
Mathematical analysis can tell us a lot of what is happening now. A great example is a Social Network Analysis Map of “@Nordstrom” I ran last Friday, February 10.
The graph represents a network of 5,293 Twitter users whose recent tweets contained “”@Nordstrom””, or who were replied to or mentioned in those tweets, taken from a data set limited to a maximum of 5,000 tweets. The network was obtained from Twitter on Friday, 10 February 2017.
There are six major types of Twitter maps produced by Social Network Analysis. One type is called a Polarized Network. This pattern emerges when two groups are very split in their opinion on an issue. Two or more dense clusters form with little interconnection. Generally one sees polarized Twitter maps for divisive issues such as women’s rights in the Arab world or a hotly contested gubernatorial race in, say, Texas.
The most effective method of reading the map is to observe how hashtags cluster within the software. With our map we see three major groups forming. Below is a summary of how the dominant hashtags in the three largest groups cluster.
It is evident that G1 are mildly anti-Trump, perhaps media and those not thrilled with the new administration. G2 are the anti-Trumps. Hashtags in G3 are, evidently, supporters of the President and Ivanka’s line of clothing. The software captures every individual tweet in an Excel file. It is possible to drill down, if the client asks, with text analytics.
The interesting finding is this: Why has Nordstrom, a chain of luxury department stores usually found in upscale malls, now become a symbol of resistance against the new administration? We think we know the answer. What is interesting is that the results show up clearly when we run a Twitter map. Is upside down the new normal?
Visualization of social networks is now coming online to make sense of Big Data and convey the results of analyses through emerging, open-source programs. This kind of analysis is not limited to Twitter, but also can be applied to other social media data, megadatasets, consumer sales data from, say, a major supermarket, Walmart or survey data. It is a great new tool that, together with our analytic skills, we can deploy to give our clients the story.
A good use of this tool is if a major brand launches a new advertising campaign. By running a Twitter network map every day for 30 days, we can gauge the penetration of the message, which hashtags are going viral over time, how they are clustering, and what is the trending message.