Observatory on Social Media

[Observatory on Social Media (OSoMe)]

You are receiving this message because you are subscribed to the Google Group that used to be called "Truthy Discussion" and that was just renamed "Observatory on Social Media (OSoMe)". This mailing list has been silent for a while but we have been busy and so we thought you'd like an update on our recent activities.

Today we released a new suite of tools allowing researchers, reporters, and the general public to access derived data from our large historical collection of (70 billion) public tweets. The suite includes an API and several visualization tools. You can learn more by reading our paper preprint (https://peerj.com/preprints/2008/), or the tl;dr version at this press release from Indiana University, that also came out today:

http://news.indiana.edu/releases/iu/2016/05/observatory-on-social-media.shtml

Our new system is called Observatory on Social Media (OSoMe in short, pronounced "awesome"), hence the change in this group's name. Please give it a try and let us know what you think! You can find it here:

http://osome.iuni.iu.edu (or if you forget, try socialmediaobservatory.org or the old truthy.indiana.edu)

In other news from our lab, we recently released an API for our BotOrNot tool to classify Twitter accounts as bots or human. It is getting a huge amount of traffic (millions hits in a short time). We recently updated our machine learning model so that its accuracy has improved (paper in preparation). This work was presented at the WWW conference in Montreal a few weeks ago (it got an award for best presentation). Articles are also coming out in June issues in both Communications of the ACM and IEEE Computer. Give BotOrNot a try, it is also under Tools on the OSoMe site (or get to it directly at http://truthy.indiana.edu/botornot/).

We are also preparing a mash-up of the OSoMe API with the BotOrNot API so that one will be able to see how much of the online conversation on a topic is sustained by bots... Find it very soon on the OSoMe website under Resources. We hope people will come up with other creative uses of our tools and APIs.

Follow us at @truthyatindiana or https://www.facebook.com/TruthyAtIndiana/

The OSoMe team

[Fil Menczer]

As a follow-up to our earlier announcement, today we released the OSoMe + BotOrNot API mashup that allows to explore whether a Twitter conversation (hashtag) is sustained by social bots:

http://osome.iuni.iu.edu/resources/
https://github.com/IUNetSci/osome-mashups/blob/master/botornot%20active%20users.ipynb

Best,

-OSoMe Team