Homework 6: Visualising Twitter

A Reddit Dystopia

A Reddit Dystopia

This Markdown exists so that we can take a look at some twitter data about the World of Warcraft guild “A Reddit Dystopia” or ARD.

ARD is a community of gamers, we spend at least 2% of our time making bad puns. We strive to make World of Warcraft as open and inviting as possible to a huge community of players around the outside world.

If you play World of Warcraft, or are interested, check us out. If you’re intimidated by the game, that’s understandable, there’s an awful lot to choke down at first.

But first, a little setup.

Obviously, we need data to work with. Fortunately, they have a twitter account - @WoW_ARD - it’s mostly used to humble-brag about raid progression

Fabulous ARD Twitter

# Download your file from the internet

# Well, twitter API verification isn't playing nice...so let's just pull the data locally.  It's not ideal, but for now it works.

# Pull Timer in 3..2..1
download.file(
  "https://www.dropbox.com/s/wv1xdtfa6do6fpd/ard_tweet.csv?dl=1",
  destfile = "Ard_Docs.csv",
  mode = "wb"
)

# Isn't that URL ugly?

# Let's import that stuff so we can work with it.

ardtext<-read.csv("Ard_Docs.csv")

We also need some libraries to make this all work.

library(ggplot2)
library(tidyverse)
library(httpuv)
library(rtweet)
library(tm)
library(SnowballC)
library(wordcloud)
library(RColorBrewer)

# This is just book-keeping, but its a good foundation with which to build any R product.

Finally, some data.

What is ARD Tweeting About?

The first thing I want to do is take a look at what ARD TWeet’s about. That’s the best way we can get a gauge on ARD’s character.

This is a Word Cloud assembled from the Tweets that ARD has sent. I genuinely expected this cloud to be more sarcastic than it is. But it speaks wonders as to the inviting nature of our little corner of the internet.

This is a “Friendship Birb”

For a time ARD ran a series of free public events, sheperding players through some fairly difficult content that they might not otherwise have access to. The bird (seen above) was a reward for completing that content that cannot be earned anymore.

Shown Another Way

# Plot the Frequent Words

barplot(d[1:10,]$freq, las = 2, names.arg = d[1:10,]$word,
        col ="lightblue", main ="Most frequent words",
        ylab = "Word frequencies")

For all of the silly memes we share, ARD mostly makes people feel good. ^So ^long ^as ^you ^play ^Horde ^{^of} ^{^course}

Hashtags for The Horde

Positive vibes and words of affirmation and all well and good. But this is Twitter after all, the only thing that matters is hashtags when it comes to figuring out how cool the kids think you are. Let’s take a look and see how ARD is doing on that (vital) front.

For the Horde indeed

Once again, we see “Friendshipbirb” popping up with great frequency, this is a big even for us. Additionally you see “battle for azeroth” mentioned. This is the most current crop of content for the game, and is the big reason that ARD is so pro-active in tweeting about Raid Progression.

A Guild is a collection of players, much like…well a team, or collection of teams. And they compete with other guilds to complete content first. In the early months of Battle for Azeroth two guilds - “Method” (from Europe) and “Limit” (from the US) were at the absolute bleeding edge of this content engaged in a neck and neck race to complete the hardest levels first.

I know…it sounds incredibly nerdy. And it totally is. BUT…the competition to do well in this arena is fierce, it is driving Twitch.tv wild and there are millions in advertising dollars in this sector, and millions more in merch floating around as well.

The folks who make this game have every incentive to keep that level of excitement as high as possible.

Who is ARD

We have a handle on what ARD does (mostly Memes)…but who is ARD?

The WoW_ARD Twitter Handle is likely not a personal one, and is shared among the leadership (it’s probably a shady cabal), but let’s see where ARD Tweets from.

ardtext %>%
  ggplot(aes(x = source)) +
  geom_bar() +
 theme(axis.text.x = element_text(angle = 90, hjust = 1))+
  xlab("Device Used")+
  ylab("Count")+
  ggtitle("What does ARD Tweet On")

ARD only uses a computer. The only possible conclusion from this data is that no one running ARD has a Cell Phone. Okay, I’m clearly joking, but it’s nice to know that there’s consistency. Let’s take a look at how “cool” ARD is, next, by taking a look at Likes and Favorites.

Conclusion

Thanks for stopping by my R-Pub. I’m having a blast learning R and trying to unlock Markdown. The tone of the article was intentionally playful, because while stats can be very dry, it also helps to make some of the most fulfilling hobbies possible.

Whether you enjoy Fantasy Football, Boxing, Competitive Poker or eSport Mythic Raiding - statistics is at the heart of all of it.

Hope to see you online, Joe

Grove