Guess the Poster CCC-WordCloud Fun Competition

[chrisw]

5. Fernando
6. Terry McCracken

Sorry! Both wrong.

[Henk]

3. Fernando for i see the words 'machine' and 'friend'

[chrisw]

3. Fernando for i see the words 'machine' and 'friend'

correct. and now the other two are logically solved by previous answers.

So it was like so:
1. hgm
2. mclane
3. fern
4. Rebel
5. Terry McCracken
6. Rolf

Fascinating, I thought, that 6. Rolf was the most proposed choice but as chrisw, bob, Ed, Vas, Dann Corbit and Terry McCracken.

I'll do another six, made slightly easier because it will contain some (randomly selected) posters who expressed an interest in this thread. Unfortunately, this means deleting the current wordclouds to make board space for the next ones.

[MikeGL]

3. Fernando for i see the words 'machine' and 'friend'

correct. and now the other two are logically solved by previous answers.

So it was like so:
1. hgm
2. mclane
3. fern
4. Rebel
5. Terry McCracken
6. Rolf

Fascinating, I thought, that 6. Rolf was the most proposed choice but as chrisw, bob, Ed, Vas, Dann Corbit and Terry McCracken.

I'll do another six, made slightly easier because it will contain some (randomly selected) posters who expressed an interest in this thread. Unfortunately, this means deleting the current wordclouds to make board space for the next ones.

Kindly double-check number 3, are you sure he's not "Father" because it has Friend and Machine. I mean "Father" the guy who
posts his fortress wins against machines which loses on time.

Kidding aside, this thread was honestly fun. Thanks for those pics. Wonder how you generated them.
I imagine scraping those forum sites then running them into an app.

[chrisw]

3. Fernando for i see the words 'machine' and 'friend'

correct. and now the other two are logically solved by previous answers.

So it was like so:
1. hgm
2. mclane
3. fern
4. Rebel
5. Terry McCracken
6. Rolf

Fascinating, I thought, that 6. Rolf was the most proposed choice but as chrisw, bob, Ed, Vas, Dann Corbit and Terry McCracken.

I'll do another six, made slightly easier because it will contain some (randomly selected) posters who expressed an interest in this thread. Unfortunately, this means deleting the current wordclouds to make board space for the next ones.

Kindly double-check number 3, are you sure he's not "Father" because it has Friend and Machine. I mean "Father" the guy who
posts his fortress wins against machines which loses on time.

Kidding aside, this thread was honestly fun. Thanks for those pics. Wonder how you generated them.
I imagine scraping those forum sites then running them into an app.

All done using Python. My youngest son is wildly outclassing me on ML and neurology front, so I decided this was not acceptable and have been playing catch up (fat chance!). One can learn a lot fast playing Kaggle things. Yes, I scraped the forums, the objective was to perform NLP sentiment analysis using a “toxic post” database, but I got diverted from that project to help Ed, and had to learn networkx stuff. Prior has been the usual image recognition things and of course chess NNs. Wordclouds are pretty easy to do, only a few lines of Python. So, all in code, no apps.

And yes, it is guaranteed Fernando, not Father.

[Sam Hull]

3. Fernando for i see the words 'machine' and 'friend'

correct. and now the other two are logically solved by previous answers.

So it was like so:
1. hgm
2. mclane
3. fern
4. Rebel
5. Terry McCracken
6. Rolf

Fascinating, I thought, that 6. Rolf was the most proposed choice but as chrisw, bob, Ed, Vas, Dann Corbit and Terry McCracken.

I'll do another six, made slightly easier because it will contain some (randomly selected) posters who expressed an interest in this thread. Unfortunately, this means deleting the current wordclouds to make board space for the next ones.

Kindly double-check number 3, are you sure he's not "Father" because it has Friend and Machine. I mean "Father" the guy who
posts his fortress wins against machines which loses on time.

Kidding aside, this thread was honestly fun. Thanks for those pics. Wonder how you generated them.
I imagine scraping those forum sites then running them into an app.

All done using Python. My youngest son is wildly outclassing me on ML and neurology front, so I decided this was not acceptable and have been playing catch up (fat chance!). One can learn a lot fast playing Kaggle things. Yes, I scraped the forums, the objective was to perform NLP sentiment analysis using a “toxic post” database, but I got diverted from that project to help Ed, and had to learn networkx stuff. Prior has been the usual image recognition things and of course chess NNs. Wordclouds are pretty easy to do, only a few lines of Python. So, all in code, no apps.

And yes, it is guaranteed Fernando, not Father.

Curious to know if you strip quoted posts of others before analysis ...

-Sam-

[chrisw]

3. Fernando for i see the words 'machine' and 'friend'

correct. and now the other two are logically solved by previous answers.

So it was like so:
1. hgm
2. mclane
3. fern
4. Rebel
5. Terry McCracken
6. Rolf

Fascinating, I thought, that 6. Rolf was the most proposed choice but as chrisw, bob, Ed, Vas, Dann Corbit and Terry McCracken.

I'll do another six, made slightly easier because it will contain some (randomly selected) posters who expressed an interest in this thread. Unfortunately, this means deleting the current wordclouds to make board space for the next ones.

Kindly double-check number 3, are you sure he's not "Father" because it has Friend and Machine. I mean "Father" the guy who
posts his fortress wins against machines which loses on time.

Kidding aside, this thread was honestly fun. Thanks for those pics. Wonder how you generated them.
I imagine scraping those forum sites then running them into an app.

All done using Python. My youngest son is wildly outclassing me on ML and neurology front, so I decided this was not acceptable and have been playing catch up (fat chance!). One can learn a lot fast playing Kaggle things. Yes, I scraped the forums, the objective was to perform NLP sentiment analysis using a “toxic post” database, but I got diverted from that project to help Ed, and had to learn networkx stuff. Prior has been the usual image recognition things and of course chess NNs. Wordclouds are pretty easy to do, only a few lines of Python. So, all in code, no apps.

And yes, it is guaranteed Fernando, not Father.

Curious to know if you strip quoted posts of others before analysis ...

-Sam-

Well, I think I do, just checked. Yes, all posts are stripped of everything except poster text. Thanks for the OMG-did-I-overlook-something-so-stupid moment.

BeautifulSoup btw.

From the stripped text, words are extracted into a list, everything is lowercased, and all non a-z characters removed for purposes of WordCloud. Actually, one only needs to look at Thorsten's worldcloud for CTF to be assured of this.

[chrisw]

Wordclouds 1-6 are about to disappear, making space for six more ...

[MikeB]

In the attachments are six WordClouds composed of texts, by individual authors, posted into all CCC forums (less CTF) since records began.

Each attachment has two WordClouds, each WordClouds labelled 1 to 6

CCC-wordcloud-1.0.png
CCC-wordcloud-2.0.png
CCC-wordcloud-3.0.png

So, competition, can you identify the poster from the wordcloud?

1.
2.
3.
4.
5.
6.

+1 - good entertainment, thank you.

[Ovyron]

I recommend using a third party image hosting service such as https://funkyimg.com (which is free) so you don't ever need to delete the images to make room for more, and future generations will be able to see them.