You are viewing a single comment's thread from:

RE: Exploratory Analysis: Statistical User Profiling for Data Driven Votes and Improved Business Performance

in #utopian-io6 years ago

Hi @paulag, seems like you hit a quite controversial topic with that analysis! From technical perspective I understand that you took a few metrics as examples to show the possibilities and much much more could be done.
While I'm "guilty" myself for applying methods like these to some extend with my other accounts, I'm a bit skeptic if Steemit or dApp stake applying data driven voting would be really a good idea. I think the metrics had to be chosen much more carefully and differentiating than the examples. Would new users join an app that rewards older accounts higher? At which point turns "loyalty" in "spam"? Gentlebot applied data driven voting to some extend, and it didn't take long until people realized at which point a simple "nice post" can give a pretty good chance for a $1.5 vote (this seems to be fixed now). Algorithms can and will be tricked, but so can humans.
I think the current set of dApps with human curation or moderation is a step into the right direction for Steem. Of course, an algorithm scales much easier with growth than human resources. I'm torn, but from my personal opinion, I think I'd prefer to add more "brain" to the process instead of replacing it with an algorithm.

Your contribution has been evaluated according to Utopian policies and guidelines, as well as a predefined set of questions pertaining to the category.

To view those questions and the relevant answers related to your post, click here.


Need help? Write a ticket on https://support.utopian.io/.
Chat with us on Discord.
[utopian-moderator]

Sort:  

yep looks like it hit a nerve, but lets face it, data is a fact of business these days.

I agree, more 'brain' would be better, but like you have pointed out, algorithms can often scale better. And yes, the metrics must be chosen carefully, and would probably differ between app depending on the business needs. I was previously working on a 'contribution score' metric and as mentioned a few times UA is on the way in some form. 1 UP as mentioned in the comments is also using data to help. I think some people and apps are going to use data to help them grow, and probably more than are willing to let on right now :-)

Thank you for your review, @crokkon!

So far this week you've reviewed 3 contributions. Keep up the good work!