Anyone have a good reference/article/suggestion on models to build automated trust/skill scores for users/forums?
Some context. I am looking at putting together an education site for professionals. But don’t want just anyone to be able to respond to certain questions, ie they need to build a trust/skill score to do so. I really don’t care how much they read or if they make comments in general, ie they can be active but might not know what they are talking about. So far the best idea or concept I like is a voting system, so when someone answers a question people can indicate if it was a good/bad/incomplete answer and weight the scores based off the existing trust level of the voter. As their score goes up they gain more access.
Appreciate any insight the community has, or if there are any Bubble Templates out there doing something like this already?
I’ve seen several models for this. One popular one is weighted based on how many responses you have posted (such as a review site). You can easily become an “outstanding contributor” by posting a lot, even though the usefulness of the posts isn’t graded or judged.
I would suggest a simple combination of:
ratio of “thumbs-up” vs. “thumbs-down” votes for the post
number of posts that have at least one “thumbs-up” vote
frequency of posting where the post got at least one “thumbs-up” vote.
This way, you can more readily identify those newer ones who are just starting to post, and they can still get a decent start on their “trustworthiness” by scoring a decent # of thumbs-up votes.
As they become more seasoned users, you can bump up their trustworthiness by the number of posts they have made that were viewed as helpful and got at least one thumbs-up vote.
The final factor is frequency of posting. If you have a genius posting once every month, their ranking should be lower than a subgenius who posts regularly and consistently gets thumbs-up votes.
One thing I would suggest staying away from is adding an additional weight factor for how User A’s trustworthiness compares to User B’s trustworthiness. Keep it based on how the community grades each user, rather than doing a comparative weighting, otherwise this runs the risk of making a less-smart user who contributes more look weaker than a smarter user who contributes less.
And increase their trust score by age of account. Don’t trust those posts from bots that sign up 1 minute ago.
Also my best advice for people who want to be disruptive on platforms is to shadow ban them. Let them feel like they are not banned, but no one else can interact with that user or their content. This prevents the disruptive user from creating another account.
