Toxicity percentage by two years of data collecting

Your reverse text generator is broken lol.

Nope. . .

gnihton laeh lliw ti ;egnever htiw egnever teem ot sselesu si tI - this however sounds like Biden lol.

89 111 117 32 97 114 101 32 97 110 32 105 100 105 111 116 32 97 110 100 32 100 101 115 101 114 118 101 32 97 108 108 32 116 104 101 32 116 111 120 105 99 32 99 111 109 109 101 110 116 115 32 121 111 117 32 103 101 116 32 108 111 108

A very trump reply.

ASCII translation of your reply is very interesting. Unlike you, I don’t insult others.

Anyway, if we’re going to talk about what you’ve been able to decipher, even on code that was written in a simple enigma.
I just want to point out that no matter how much this post looks like a fake, it will at least bring down one of the biggest communities that does boosting etc. on European servers.
This is already known and people from our group are working on it.
If Blizzard doesn’t want to address this problem we will have to take care of it ourselves. We already have some ways to do this and it is more than likely that they will be banned.
If only because in the current global situation they are breaking the laws of many countries or indirectly forcing their customers to risk their own safety regarding the laws.

As for the context of the analysis of the forum itself.
Many of you will know how and why we got the information through the forum articles themselves, but it must be added that a large part of it was just a part of the in-game community.

We have evidence from discord server conversations.
We have the bank account numbers where the funds and earnings of the ToS violating communities went.

We have pretty much everything we need. We’re just missing a tiny bit of information against what we’ve already found to put a full stop to their activities.

Anyway, the overall chart with all the information attached by all the sources and the evidence of the toxicity itself in not only EU but US communities is the main work that has gotten us to sink the big money making ship.

Such results will be presented before we act.

The toxicity was not only about how and who was able to influence public opinion on the forums and thus contribute to the improvement of the game itself but also how and why the suggestions for change that the community is now grumbling about the most were brought to the forefront.
Some of the members of such communities immediately influenced people’s opinions in several ways. (The communities I’ve already mentioned.)

1 Like

You know what’s sad - I can fully believe that figure.

1 Like

So we’re back after the second flag. I guess someone is having too much fun :slight_smile: .

Blizzard will ban you fellas, for ruining their business.
They are passive about it for a reason.

lmao be more naive pls

naive aint the right wording.
Realistic maybe?
I support what you doing. Good luck.

My Demon Hunter (who has been quite agreeable with your posts in the past) would be in trouble if this was in any way legitimate. It won’t be though, as this just looks like stirring the pot to me.

Blizzard cannot ban you for out-of-game activities and your personal life on social media or non-Blizzard platforms.

Ban aint the right wording also, I must say.
Anyway if its true what you say, good luck. Fingers crossed your project will deliver results.

1 Like

I can safely recommend anyone who has ever done boosting activity or anything similar to stop doing it immediately and delete everything associated with it. Permanent bans will be handed out. There is only one thing missing and that is the name of one person. As soon as we get it through legal means these communities from big to small will fall. It will affect the discord itself and other communication platforms. Plus the worst part is most of the money from these activities is on the border of the laws of several EU countries immediately due to the current situation IRL, at least as far as money movement is concerned. Unfortunately without that one name we can’t do anything. It would be naive to think he will speak up on his own. Anyway, we probably already know his nickname in the game. That’s why we want to take our time.
If this works out, it will be one of the biggest cases ever solved in history involving a computer game, and I have to admit that we got to this point quite by accident thanks to the forum and the statistics we’ve been building.
Furthermore, it would be great to get around such cases further, preferably by Blizzard changing the rules of the forum itself by removing the ability to switch between characters on the forum.
I understand that Blizzard itself has access to information about who is what character, but it has been hell for us to find out who is what character.

I reckon this has something to do with IRL money laundering?
that honestly sounds like the only real thing you could accuse such a community of.

Also if you don’t have the ball securely in the net;

why are you posting this now?

I would bet on- tax avoidance.

What is exactly the method used for collecting and annotating that data ? For example, who and how was it decided if a comment was positive or negative ? Have you used something like an AI that would automatically do the job, without human interference (but then it would be prone to errors as its F1-score would be less than 1.0) or have you used the partial views of a particular human or the partial views of a group of humans ?

It’s already difficult to classify an object from a dataset when this object has ambiguous properties, and here you’re talking about classifying objects which are by nature ambiguous, just because of the non-context free nature of any natural language, without even taking in account things like sarcasm, humor, etc…

So… I’m a bit interested… And don’t be afraid to go technical. I have a PhD in machine learning and NLP.

2 Likes

People. You can find the table according to what and how we evaluated the comments in the results when they come out. But it’s nothing complicated.
All were voted on and discussed. Some contributions were excluded from principle (they could have been thought of differently or the profile that wrote them would have contradicted itself in its opinions

Yea, that’s pretty much my question though. How were they voted and discussed ? What were the criterias and how it was decided if a particular comment met a criteria or not ?

You’re also talking about “profile that wrote them would have contradicted itself in its opinions” how these opinions were fixed on that particular profile ?
Have you build profiles for each commenter in your dataset ? How was these profiles made ? Have you tried to extract the usual topics each profile likes to engage into, for example ?

1 Like