Hey lads. Out of curiosity and practice, I made a program that checks how much different cunts discuss subjects here on Zig Forums. It's very simple and has probably been done before, but it was a good exercise so idc. What it does is search for specific keywords on the archive with the different countries, and calculates how much more common that topic is within the different countries compared to outside. Then it spits out the results in a self-contained chart. The search keyword is also included on the chart if you think I'm making stuff up.
I'll be dumping some of the more interesting charts I've already made. Keep in mind that 1 is about the average. Also you can suggest some different topics/buzzwords I can try to check later, but I'm more likely to do it if you give me a proper search string that works on desuarchive.
Couple things to keep in mind:
- Smaller countries have smaller sample sizes thus more outliers, so the extremes will be dominated by them. Could filter them out, but it might be interesting so I won't. You could focus on the countries with more posts (see next chart) if you want some proper data.
- This doesn't consider the language used in the post, so there can be language clashes in that words mean something different in other languages. I'll usually check the highest results for this, but the onus is on the country to explain their results.
- Whether the post is negative or positive about the topic is not registered, so you can instead see it as how OBSESSED the cunts are with the different topics.
Zig Forums topic prevalence revealer
Other urls found in this thread:
Here, by the way, is the percentage of Zig Forums posts by each country as I've found it.
Just a couple of dry geopolitics ones first
Post the results for:
*race
*white
*black
*latino
Searched individually or together? I think individually would be too uncertain
yeah, nevermind actually
Already made it though
based my senpai
keep up the good work i hope you make this a regular thing
are you currently just using everything in archive as data ? have you considered doing this month by month or week by week ?
also there's the language problem as i can see by how underrepresented my countrymen are in the anime category
but you are right and white and black could just be people talking about the literal colors
The one youve been waiting for
lose weight
also i see another problem. some flags are used 99% by VPNfags and end up high on the relative graphs, you could maybe disregard some of those.
we had a guy on /isr/ who kept switching between different rare flags every post
It seems like every one of these is the same
I'll post some porn related ones a bit later
all me 2bh
really though me and the other hk flag who posts on this board don't deserve to be represented on statistical graphs
It's all just showing who posts the most on Zig Forums. USA will always have the most and will usually be followed by UK or Canada.
Using the total, though it would be easy to restrict it to the last month or whatever if needs be. Could be interesting to examine some recent behavior.
And yeah the language problem is a big problem, but this is admittedly a very simple program. I would need to examine every post individually to fix that, but could be done over a shorter interval.
On Zig Forums that would really be drowned out by all the racebaiting, imo
could you make the inverse ? like check by country and list the most commonly used words (outside obvious shit like 'the','if' and 'and')
i wanna see the most common words by israelis
can you somehow *are ____ white*?
you could use
translatr.varunmalhotra.xyz
to translate every word you have into a gorillion langauges and put that in your query
might get some mistakes but its probably mostly good
Yeah, I think one should mostly just focus on the large countries. Might post versions with only the most trafficked countries later. I keep the smaller ones as a novelty.
You should look at the bar chart. The pie chart will always show the most avid posters first.
Yeah, the smaller traffic countries are mostly there for novelty
Shouldn't be so difficult, but will be a lot more demanding. Perhaps if I restrict it to the most recent posts.
why's the uk missing?
awesome work OP, thank you!
reckon either 1. are lower than austria or 2. more prone to talk about the news network so mess with the numbers
Here you go. Excluded "are not white" because it's from a completely different context, imo
I may do that, actually. I also try to restrict myself to more universal terms Thanks for the feedback, though. Another thing is to just to try to ascertain the language of each post, and remove it if it's not in English.