Jump to content

New strategies to filter bad words


Recommended Posts

The current profanity filter is fine but could be fine-tuned because it can be skipped in many ways, for example by typing "fu-ck", "f-uck" ... etc, also, if a word in some language includes a chunk of a bad word is hidden, for example, "enfadarse" (get angry in spanish) is filtered and shown as "enfad ****".

 

Based on another software to filter bad words the strategies could be:

 

  • Partial match, will flag a text as profane if any substrings of it is in a dictionary.
  • Allow symbols, will flag a text as profane if any word in the text matches a dictionary after removing the symbols.
  • Duplicated characters, will flag a text as profane if any word in the text matches a dictionary after removing duplications.

 

An example software (written in Ruby) https://github.com/cardinalblue/profanity-filter

Edited by bellzebu
Link to post
Share on other sites

"skipped in many ways" not really. All is saved in log. For now we are simply closing eyes on it...

one thing is when you write "fuck" and this word is hidden and much worse what you can do is try to get around BWF by writing f u c k, or fu-ck. In this case punishment will be guaranteed mute and in some cases even ban.

  • Thanks 1
Link to post
Share on other sites

As OMA has implied, all of this type of stuff was thought of beforehand. Spacing out words and other common tricks will still be flagged by the BWF. The only reason such things are not ******'ed in chat is to avoid random false positives.

 

You may be surprised to know that most people do not even bother... 🙄

 

Example:

image.png.632a8f1b0b9519fc2100de0e52d5736d.png

 

(^^ OMA feel free to delete this if you don't like it posted here 🙂 )

Edited by impossybull
  • Thanks 1
Link to post
Share on other sites

My original proposal was not focused on preventing bypassing the filter but on improving strategies to avoid precisely those false positives, as in the original example, "enfadarse" is not a bad word and is marked as such "enfad****".

  • Like 1
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.