Getting Started Create a character Create an event Create a planet Create a ship Posting fan fiction Image uploading Fanon Good Articles Featured Articles Characters Galactic Republic Imperials Jedi Mandalorians Separatists Sith Read more Events Assassinations Battles Bombings Campaigns Holidays Wars Read more Organizations Companies Criminals Force-based Governments Read more Planets Core Deep Core Inner Rim Outer Rim Unknown Regions Read more Read more Fan Fiction Featured Fiction Featured Fiction library Nominate a story Requirements Novellas Novels Short Stories Read more Community About Star Wars Fanon Café Fanfic Community Forum Writing Center Fanon and Fan Fiction Announcements Star Wars Fanon Discussion Consensus Track Guidelines Article Guidelines Image Policy Editing Guidelines Read more Live Chat Chat Guidelines Join Chat Tutorials Image uploading Posting fan fiction Writing featured articles Read more
119 Posts

Automated FWN and FAN screening

I'm excited to finally announce that I'm very close to an initial release date for a special project of mine that is near and dear to my heart. It's called the Automated Transcription Analysis Review Unit, and it's a bot. Unlike other python-scripted bots, it's not designed to clean up the wiki as a whole--it's designed to conduct a very basic analysis of articles and stories.

How does it work? It's simple. You enter the page that you want it to evaluate, and the bot does a quick analysis of glaring issues across the linked page and all subpages of that article. I designed it to help speed along Featured Work Nominations, but I'm hopeful it can be applied to Featured Article Nominations.

Current Discriminators
  • All missing categories
  • Partial missing eras tags (won't catch on subpages)
  • Ridiculous numbers of commas per 1000 words
  • Individual common spelling errors (e.g. "hanger"/"hangar")
  • Ludicrous numbers of superlatives (e.g. "harder," "better," "faster", "stronger") per 1000 words
  • Far too much violence (filters on words like "blood," "spewed," "gore," and various anatomical references) per 1000 words
  • Overuse of profanity (haha, like I'd say what it filters on here)
  • Overuse of sexual references (see above)
  • Lack of quotation marks (Missing quotation marks around words like "said," "told," "described,") per 1000 words
  • Scenes are too short/lack development (size of prose blocks between scene or chapter breaks)
  • Sufficient description/detail (searches between line breaks for sections with quotation marks. This was hard to do, but essential.)
  • Way too much Mando'a.
  • Fewer than 500 words in the BTS section
Planned discriminators
(These are future upgrades I have to figure out how to include logically in the bot, but would be great enhancements).
  • Large numbers of "it's"/"its" misuse
  • Overuse of repeated words (excluding pronouns, articles, conjunctions, and linking verbs)
  • Overuse of Mandalorians
  • Overuse of clone troopers
  • Overuse of Revan
  • Like, seriously, overuse of Revan
  • Gunship or Warship weaponry per 100s of meters in size
  • Unacceptable Mary Sues
  • Yeah, actually, all Mary Sues.
  • Serious overuse of superweapons
Known Issues
  • No PDFs! Absolutely none! The bot can't parse them.
  • The bot's personality can be a bit…prickly…at times. The machine learning tools I was experimenting with apparently were influenced by my own online persona. That's what happens when you teach it how to evaluate based on my FWN evaluations and objections and MPK's rants, I suppose.
Testing

Please feel free to try out the bot below! Let me know what you think.

Bot

EV9D9-ROTJ
Beep Boop! I am a bot!

My purpose is to do a quick and painless evaluation on stories and featured work nominations to lessen Ataru's workload.

To use the bot, type {{Atarubot|the name of your work}} in a comment below, meatbag

Example:

{{Atarubot|The Final Judgment}}

Enjoy!


0 11
  • Upvote
  • Reply

This is really exciting! I'm glad everyone now has a quick way to check out the quality of their works! Speaking of, let me throw my most recent piece into the A.T.A.R.U.

Evaluation Results

Your work, One of You, was evaluated and its quality was found to be on par with The Real Empire Saga, much lower than featured work standards. Please go read a quality work like this one or review the guidelines and try again.

Thank you and have a nice day.

-A.T.A.R.U. Bot


What the kriff? Clearly the A.T.A.R.U. doesn't understand my genius. Maybe it'll realize my brilliance if I run one of my featured articles through it.

Evaluation Results

Your work, Nemo Qel-Droma, was evaluated and its quality was found to be on par with Jar Jar Binks, much lower than featured work standards. Please go read a quality work like this one or review the guidelines and try again.

Thank you and have a nice day.

-A.T.A.R.U. Bot


Well, the tendency to repeat itself aside, the bot isn't designed to stroke egos. I ran into technical issues with repeating itself when the old forum code I was using broke thanks to Message Wall, but I assure you, the metrics it uses to evaluate entries are quite solid. Perhaps the problem is on your end?

Evaluation of our newest FW isn't looking too good.

Evaluation Results

Your work, The Final Judgment, was evaluated and its quality was found to be on par with Darth Hatrex, much lower than featured work standards. Please go read a quality work like this one or review the guidelines and try again.

Thank you and have a nice day.

-A.T.A.R.U. Bot


Also, any Mando'a is too much Mando'a. Wink

Firedance wrote: Evaluation of our newest FW isn't looking too good.

Evaluation Results

Your work, The Final Judgment, was evaluated and its quality was found to be on par with Rawq Hahrdplays, much lower than featured work standards. Please go read a quality work like this one or review the guidelines and try again.

Thank you and have a nice day.

-A.T.A.R.U. Bot


Hmm, must be a bug in the bot's code! Next thing you know, it'll be spelling out hidden messages in the initial post. ;)

0

Firedance wrote: Evaluation of our newest FW isn't looking too good.

Evaluation Results

Your work, The Final Judgment, was evaluated and its quality was found to be on par with Darth Hatrex, much lower than featured work standards. Please go read a quality work like this one or review the guidelines and try again.

Thank you and have a nice day.

-A.T.A.R.U. Bot


I knew things weren't looking good for me when I saw "Yeah, actually, all Mary Sues"...

This bot is programmed to be about as merciful as MPK of old was. And since I was never overly fond of The Final Judgment, consider the bot's judgment…final. *sunglasses*

I have reviewed this bot and have found it to be in compliance with the Terms of Use. '''Approved'''.

Of course it is, Brandon. Not only that, but it's also Ataru finally finding his community spirit. Hopefully, this is just the beginning for Ataru.