That’s a good idea, I’ll take it under advisement. But for now I’m not going to implement it because I’d like to get the main and most important features first. It doesn’t hurt to get multiple answers Thanks for idea though!
Feature idea: swearing filter, which can automatically flag posts as inappropriate if inappropriate words are found in a post.
I mean… I don’t know. Not sure if that is 100% necessary because as long as you aren’t dropping f bombs, some swearing doesn’t hurt.
Also I will need some forum regulars to help me train a machine learning algorithm for various things. Please send me DM if I can count on you
I won’t probably ask you to do anything for some time because there is a lot of other things I need to do first.
Not such a big fan of automated flagging, I just feel like a lot could go wrong…
Indeed, I am thinking that flagging could also be handled by ML with supervised learning, meaning that posts aren’t flagged unless the action is approved by a bot mod.
Woah this is gonna be awesome, can’t wait to see the machine learning doing its thing.
I might have missed it, but has Glitch gotten back to you yet about letting it roam free?
Haven’t got back to my email, don’t think they have for @ihack2712’s either.
Glitch only replies to emails within business hours MON-fri, I’m still waiting for a reply. I dont want to add any features before I Get the permission to run the bot
Hi all! Just letting you know that today is a new week and Glitch is getting back to work today (remotely of course!). I’m thinking they’ll probably use 1-3 days to reply, but please note that it might take more time Have a nice day
Just a quick form to get started on ML for Automator. Anyone over the age of 13 can participate.
Over the age of thirteen? My birthday is next next week, can an exception be made?
Sorry, I was a bit unclear on that one, I meant 13 years or older This has to do with GDPR.
Dude, the third part of the test was more stressful than a Mental Ability test. And I loved the example about a nude-person-website being a custom domain removal.
how is the ml gonna work
The automated message responses will use a machine learning method called kNN with reinforcement. It’ll take these (but not limited to) factors into considerations:
- Word count
- Words used in the topic
- Whether or not the topic includes a domain of any kind.
- Whether or not the topic is asking a question (3rd party service will determine this).
- Whether or not the topic is implying that the OP wants to do something (3rd party service will determine this).
More properties will be figured out later. I’ll have to discuss this with a friend of mine who knows a little bit about machine learning.
Um. These answers are yes/no questions which basically everyone could agree in, I feel like the time making the google form coulda been sent just straight up categorizing the msgs