This is an exciting season to be a console gamer. Regardless of whether you decide to make a day-one purchase of the PS4 or Xbox One, you likely are slurping up console news and unboxing videos in anticipation of release. Perhaps, even, in a momentary lapse of 140 characters, you fired a tweet off to your devoted followers, gushing with enthusiasm for the new Kinect. Maybe you then lost half of your followers (and gamer cred), so you deleted the tweet.
Don’t worry. After the jump, we have you covered.
The face of Twitter is deceptively simple. Take this tweet, for example:
my PS4 come in tomorrow!! yes boss. – JR_LegitiFYE
What do we know about JR_LegitFYE? Bad grammar? Would you click on this tweet to learn more about its author? What if I told you that he’s a Nigerian American, living in Atlanta, who owns his own clothing line. His real name is Damola Adeoti Jr, he has 1185 Twitter followers, has written 63895 tweets, he apparently works for a photography studio, and he used the Twitter website to write that tweet at 10:51pm on November 13th, 2013.
The fun part is that all I had to do to have that information sent to my computer was to click a few boxes on the Twitter website and write a few lines of Python code.
The even more fun part is that Twitter sends me tweets related to the Xbox One, PS4, and Wii U at a rate of about 200,000 per day, and with each tweet, it sends me loads of metadata about the tweet’s author, connections, and context.
Have a look at this Gist to see what Twitter sent me about JR_LegitiFYE, just for fun. In terms of the amount of metadata that comes with most tweets, this particular tweet is very lean because it contains no hashtags, does not mention other users, and is not geo-tagged. Here is another example that contains more metadata.
Many companies now track and mine Twitter to help understand how consumers react to new product introductions, who has influential voice in the social media space, and whom to engage to increase the reach of their marketing messages. Mining Twitter is big business.
I thought it might be fun to conduct our own little Twitter mining experiment during this console-release season. On the evening of the 10th of November, I launched a script to capture the stream of tweets that mention the following terms:
xbox one, xb1, #xb1, xbox, #xbox, #xbone, xbone, #xbox1, xbox1, playstation 4, ps4, ps 4, #ps4, playstation, #playstation, wiiu, #wiiu
As of the evening of the 15th of November, Twitter has pushed 950,000 (English language) Tweets about those topics to my computer and I’ve stored subsets of their data in CSV files of 50,000 tweets each. Through the holiday season, I’ll use a variety of tools to mine the data in those tweets and I’ll report back here about the trends within them.
For those of you interested in learning how to do this yourself, I will post code and methodology in Qt3’s Hardware and Technical stuff forum. In a few quick clicks, you can launch your own Twitter mining project to follow along. If you have any thoughts about the project or what to explore in the data, please leave them in the comments or in the forum.
Happy Console Release Season and Good Tweeting!
This post brought to you by Clay Heaton, former game developer, long time Qt3 reader and forum member, founder of the nascent GameAid, and current MSc student in Advanced Analytics at North Carolina State University.