Social Graph as Mirror in the Net

in psychology •  7 years ago 

Does anyone need an online mirror? Does anyone care about their own image when posting, commenting, voting and liking online? Does anyone think it might be useful to learn more about people around, about the strengths of connections with them and about similarity with them in respect to personal interests?

If answer is — yes, then — https://aigents.com/ service can be used for this and for more for social networks such as Steemit, Golos, Facebook, Google+ and VKontakte. Here is what you can do about this.

First, you register with https://aigents.com/ with either of Facebook, Google+ or VKontakte or using your regular email. Next, you may edit your personal “Settings” and add your ids for Steemit and Golos.

Once you are done with this, just got to “Graph” menu item. Once you get there, you see “sample” social graph of your own which presents how such graphs are done.

Circle in the middle is representing yourself and circles around are your social connections. On the very top, with saturated yellow color, are ones with greatest reputation and social capital from your perspective — so these people are getting a lot of your likes/votes and comments by pay you back a little. On the left and on the right, with medium yellow saturation there are people with similar level or social capital/reputation with you — so you exchange with them with likes/votes and comments mutually. At the bottom, with low yellow saturation, there are people who provide you with many likes/votes and comments while you do not pay them back often enough.

As it is shown on example, topmost people on graph can be thought as “authorities” or “those who I may most attention to but they don’t notice me”. Below, between them and you, there are “opinion leaders” or “those who I pay more attention to than they pay to me”. At the very your level, there are “friends” and “colleagues” who you communicate with symmetrically. Below you and the very bottom, there are “followers” or “those who pay more attention to me than I pay to them”. Finally, at the very very bottom, there are “fans”, who pay a lot of attention to you but you barely return it to them.

What we call social capital or reputation here, rendered as vertical position of a person and saturation of yellow color can be also called “karma”. Note, the calculation of this value may be imprecise because only communications specific to your own news feeds are accounted and don’t involve communications in other groups or feeds of your connections.

Communication connectivity between you and people around you is represented with blue arrows. Relative lengths of arrow between you and person indicates ration of incoming and outgoing actions — likes/votes and comments. Respectively, for people above you, links from your direction to them are more long. In turn, links from you to people below are shorter. Moreover, width of these arrows indicate relative intensity of communications with given person compared to other partners on the graph.

Another feature of social graph is indication of similarity of people in respect to topics of interest — this is indicated as size of light blue halo around each circle.

To get real use of it, just click on one of the social network images in the right-top corner — you should some of them enabled there, based on social networks you registered with (Facebook, Google+ or VKontakte) or those that you linked in your personal “Settings” (Steemit or Golos).

There, you will get the real graph of your won and can view it with few more options, available with few widgets in the left-top corner.

First, you can restrict rendered social connections by level of similarity with yourself. By default, no filter is set by you can change it to see only people similar as 25%, 50% or 75% or more.

Next, you can restrict rendered social connections by level of connectivity or intensity of communications with yourself. By default, no filter is set by you can change it to see only people with connectivity as 25%, 50% or 75% or more.

Finally, you can specify period that you want your graph to be rendered for — 1 day, 1 week, 1 month, 1 quarter, 1 year or all years online.

And one more feature is available for public social networks based on block-chain technology such as Steemit and Golos. You can actually browse entire social graph on the network, clicking on any node on the graph and changing your view focus to other person.

The more is coming, graphs of interests and sentiment are coming as well as option to track development of your own social capital reputation (AKA “karma”) in historical perspective.

Stay tuned with us:
https://aigents.com
https://www.youtube.com/aigents
https://www.facebook.com/aigents
https://plus.google.com/+Aigents
https://medium.com/@aigents
https://steemit.com/@aigents
https://golos.io/@aigents
https://vk.com/aigents

Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!
Sort Order:  

Hi! Found your article via Medium and was happily surprised it was related / mentioning Steemit! (Was looking for articles on privacy in social networks)

Looks like a really interesting service you made, def gonna try it out later today! All the best!

Just been able to test it out, interesting to see really! And looks quite accurate as well, well done!

Just curious, what metrics are used to evaluate similarity ?
And is level of connectivity measured by public posts, or does it account for private messages (messenger on Facebook for example) as well?

Looking forward to discover more of your projects @akolonin !

  ·  7 years ago (edited)

Thank you @lauraorigin! In Facebook and Google+ analysis involves only information accessible via public API-s of those services, which limits only communications (posts, comments and likes/votes) in home pages of the themselves users and does not count groups or home pages of the other users. In Steemit and Golos, almost everything is counted (we don't handle reposts for the moment but we are working on this). Also we are planning to do sentiment extraction everywhere in the close future. Here you can find more detailed information on what we are doing https://aigents.com/papers/2017/Assessment-of-personal-environments-in-social-networks-2017-IEEE.pdf . Shortly, similarity is calculated based on words that people are using but we are going to change it to use tags (if present) too. Most frequent updates are happening on https://www.facebook.com/aigents . Thank you one more time :-)