Introducing UserAuthority (UA), @steem-ua and UA-API !

Introducing UserAuthority (UA), @steem-ua and UA-API !

steem-ua.png

Great news! After a lot of research and months of development by @holger80 and myself (@scipio), we finally made it happen: UA is now a reality!

Introducing UA: a new kind of Reputation Score

For those of you that haven't heard about UserAuthority (UA) already, I introduced it as an algorithm in my article How-to solve SPAM and Democratize Steem: Introducing UserAuthority in November 2017. In essence, UA can be seen as a new Reputation metric: it looks at the entire Steem Follower Graph and carefully tracks who follows who. It doesn't directly care about how "rich" an account is, or how old it is, or how much an account has posted, but please keep in mind that those factors do indeed indirectly influence any account's UA score.

The currently used Reputation metric used by Steem itself, can be "bought for": to put it simple, "Rep" looks at the total amount of "upvote value" a given account has received over time. However, by self-promoting a few published posts, any user could quickly increase its own "Rep Score", which doesn't really reflect that account's True Reputation. We had the idea to develop a Score that better reflects an account's True Reputation: UA is that new score!

How are UA scores computed?

All UA scores are computed daily by first updating the Follower Matrix; we do that by continuously tracking every "follow operation" (= an account following or unfollowing another account) using @holger80's own Beem API for the Steem blockchain. With every new block produced (every 3 seconds), so is the UA Follower Matrix. And all those "follow operations" (~ 150 million currently) for all accounts on the Steem blockchain form a gigantic data cluster of ~ 1.2 trillion EigenValues. We then look at Trusted Accounts to initiate our sub-algorithm "WitnessRank" with: each day, we check all witnesses for being -a- active and -b- up to date, and -c- their Witness Stake; we use that Witness Stake to propagate "Trust" from each witness to the accounts it follows, and who they follow, and who they follow, etc. etc. etc. And after heavy-heavy HEAVY number crunching, as a result, the UA score for each account comes out.

You can check the current UA Top 100 Rank and your own current UA Score & Rank on our website https://steem-ua.com/ (for your own UA Score & Rank, please log-in via SteemConnect)!

How to increase your own UA score?

To put it simple: by being interesting / "attractive" to others. Either you publish interesting content that people want to read and comment on (like @paulag for example), or you own a lot of Steem / SP that you might want to share with the people you interact with (like @thejohalfiles for example), or you do other noteworthy things with that people might be interested in (you form a community like @dtube, @utopian-io, @steemhunt or publish an app like @steempress for example). Being a Top Witness helps as well! ;-)

Introducing @steem-ua: Algorithmic Curation via UA Scores

UA, as a Reputation / Influence metric, has quite some interesting applications. @steem-ua utilizes one of them: since every Author Account has a UA score, so does every Post Voter. Instead of looking at the SP a Post Voter owns, @steem-ua, our own Algorithmic Curation Service looks at the UA score of the Author and all its Post Voters. Using both metrics, we're able to Algorithmically Curate via the formula:

UA_Vote = x * UA_Author + y * UA_Post

where x and y are the weights we use, and UA_Post = sum(Voting_Strength * UA_Voter)

... and that works pretty good! Although @steem-ua's Curation Mechanism is completely algorithmical, it incorporates other users' votes as well.

What's the goal of @steem-ua?

With @steem-ua we want to encourage all Steem Authors to write High-Quality Content about their expertise, which could be about any topic. The higher your UA Account Score, and the higher your Post UA, the higher our upvote! Please join our Discord server and we'll gladly help you on improving your UA Scores / Rank and build up your account!

How to receive @steem-ua upvotes?

Since all types of content are welcome on @steem-ua, and because we cannot upvote every possible Post published via the Steem Blockchain, we decided to let the @steem-ua Bot score & upvote posts published by @steem-ua SP delegators:

SP DelegationSponsor Benefits & Upvote Frequency
25 SPmax. 1 @steem-ua upvote per week
50 SPmax. 2 @steem-ua upvotes per week
100 SPmax. 4 @steem-ua upvotes per week
250 SPmax. 7 @steem-ua upvotes per week
>= 5,000 SPStake-Based Sponsor Rewards + max. 7 @steem-ua upvotes per week

PS1: delegation requirements could be subject to change, we'll then post about that prior to changes being applied.
PS2: @steem-ua reserves the right to reject upvoting accounts demonstrating abusive behavior.

Nota bene: regarding your Post upvote, it doesn't matter how much you delegate to @steem-ua, every Post is scored & upvoted by its UA score: a higher @steem-ua upvote cannot be bought-for. High-SP Sponsors (with a minimum of 5,000 SP delegation to @steem-ua) do receive Sponsor Benefits, but those are based on their Sponsorship Stake using an additional upvote mechanism.

New to Steem? Don't have enough SP available to delegate to @steem-ua?

Not a problem! Just talk to us on our Discord server and we'll work something out to get you started, you're welcome!

Introducing UA-API: our UA Data Providing Service

Do you manage a Steem dApp, Community, or other type of Steem Service and you'd like to use UA Data to better manage User Access and Reputation Scoring? Now you can! In order to incorporate UA into your own service, we've created a UA-API Data Service that Steem services can use as Registered UA Partner with a Private Key. Everybody is invited! Just talk to us (@scipio and @holger80) on our Discord server and we'll help you integrate UA!

UA, Steem User Authority: Use Your Influence!

Visit our steem-ua.com website - View your UA score

Quick Delegation Links:
25 SP | 50 SP | 100 SP | 250 SP

H2
H3
H4
3 columns
2 columns
1 column
Join the conversation now