Tag Archives: spam

Is @th3j35t3r a teenaged male pop star?

securityThe Jester, also known as @th3j35t3r on Twitter and JΞSTΞR on his blog, is an anonymous political “hacktivist”. Justin Bieber is a 19-year-old Canadian pop singer. Are they the same person?

The theory

When @th3j35t3r first came on the scene in 2010, I thought he was an interesting character. However, his tendencies toward ignorant and homophobic speech, including using the word “gay” as a bad word, reminded me of a juvenile male script kiddie.

the jester th3j35t3r homophobe
@th3j35t3r circa 2010

Since then, his word usage has matured a little, and he even admitted in 2012 to not be anti-gay.

Being gay is fine. Being bi-sexual is fine. Being straight is fine. Do you know what’s not fine? Crocs.
@th3j35t3r (4:43 AM Nov 8th, 2012)

I started thinking about who else has recently matured past his early teens. I wondered if The Jester was, secretly, the teenaged male pop star, Justin Bieber.

The analysis

I downloaded as many tweets of The Jester and Justin Bieber that the Twitter API allowed me to. Although Twitter’s API says that I can download up to 3500 tweets, I wasn’t actually able to pull that many.

Even though Justin Bieber had made way more tweets, I had a smaller sample from him, so I adjusted the one from The Jester to have the same sample size, for a comparable time period.

I then extracted a common dictionary weighted by word counts, and plotted the data into a digital fingerprint of their word usage.

Although I had used systems like this in the past for digitally fingerprinting spam e-mail messages, I needed to confirm that this new simpler system might work as expected. For a control, I split the tweets from The Jester into two data sets and built a digital fingerprint for each dataset for comparison.

th3j35t3r word usage 1
Dataset I
th3j35t3r dataset 2
Dataset II

For comparison, we look at the peeks and valleys, and the majority of them line up. This indicates a reasonable success in our method at digitally fingerprinting a person by their word usage.

th3j35t3r word usage control digital fingerprint
Control comparison

I returned to my original dataset, and made another digital finger print for The Jester, whose word usage was all over the board, but who always used the correctly capitalized pronoun, “I”:

th3j35t3r word usage digital fingerprint
The Jester

Analyzing the word usage of Justin Bieber, resulted in heavy usage of the words “love”, “thanks”, and ASCII smilies, and he often didn’t capitalize his usage of the pronoun “i”.

justin bieber word usage digital fingerprint
Justin Bieber

The conclusion

th3j35t3rbieber justin bieber the jester digital fingerprint ocmparison
The Jester is not the Bieber

Although The Jester might in fact be a teenaged male pop star, it is only 5% likely that he is  the one named Justin Bieber. However, it’s also unlikely that Justin Bieber makes his own tweets, which would have thrown a monkey wrench into my analysis.

Discuss.

UPDATE: My wife has pointed out that Justin Bieber has probably never publicly communicated like an ignorant teenaged boy. No offense to Mr. Bieber or his PR team.