Recently when I was browsing online, I came across this chart. The graph depicts breakup frequency gathered by using data collected from Facebook relationship statuses. The graph kind of reminds me of something like a Google Analytics graph that charts the number of visitors over time and allows you to see trends in data.
There are some very interesting features labeled on this graph. For example, breakups are most frequent during spring break and a few weeks before Christmas. However, Christmas Day is the lowest point on the graph. Also, it’s really interesting to note how Monday appears to be the most popular day for break ups to occur.
I thought this graph was a unique example of how big data can be used to expose interesting information. The production of this chart did not require any experiments or surveys, just data about relationship statuses that Facebook tracks. On one hand, this is cool and shows some of the potential Big Data could have. Without Big Data, this graph would have been much more difficult to produce as accurately. Surveys would have had to be distributed to people and these people would have had to fill them out to from memory, which may not be exactly accurate. But with this data from Facebook, it is easy to track exactly when breakups happen.
On the other hand though, it is more than a little creepy to think about how much data Facebook has and what other metrics they could track. This example is fun and entertaining, but Facebook has so much data, you can never really know what they are keeping track of.