Yesterday we presented our paper Community Core Detection in Twitter - a “Bottom Up” Heuristic. at Sunbelt Conference in Riva del Garda. Got a lot of valuable feedback. Thanks!

Time to publish our paper on:
Community Core Detection in Twitter - a “Bottom Up” Heuristic

In this paper we present a very lightweight heuristic for detecting cores of expert communities within twitter. The heuristic combines simple text search methods with social network analysis. One big advantage of this heuristic is that it needs not to be run over the whole network. As a “bottom up” approach it explores the network around seed accounts and detects communities with simple measures.

in our recently published tool twitter list explorer (tlx) beta we don’t used that heuristic as we simply read in twitter lists, which may be seen as manually extracted community cores (which turns out not to be true in many cases, because there are several reasons for collecting twitter accounts in a list).
Anyhow it is the plan to integrate such a heuristic in one of the next versions.

The paper will be presented at the next sunbelt conference 2010 in Italy.

We proudly annonce twitter list explorer (tlx) beta

The “twitter network explorer” lets you visualize and explore Twitter lists:

• Enter any Twitter list URL, e.g. @twitter/team
• Hit “Explore” button
• Depending on size and popularity of the list loading may take up to one minute
• In some cases the network doesn’t load properly; therefore try several times, inlcuding “Shift+Reload”. (Remember: it’s “beta” – and Java
• After loading, please read the information in the “Network” tab.

Extract from our research, done with pajek and self collected twitter data:

twitter team list from here:

http://twitter.com/twitter/team

extracted list manually:[twitter_team_list]

the lists consists out of 121 accounts

i exctract the 1 all neigbourhood and get a network consisting out of 5391nodes.

here are the top 40s - all degrees

Rank  Vertex  Cluster     Id
——————————–
1    1768     1011     ev
2    1006      861     biz
3    2016      393     mashable
4    5159      387     twitter
5    2463      291     delbius
6     830      289     al3x
7    2042      245     caroline
8     741      216     scobleizer
9    1278      207     nk
10    4740      199     charles
11     940      196     jess
12    1342      180     laura
13      50      163     jack
14    4419      163     techcrunch
15    4121      157     twitter_tips
16     462      156     twitterapi
17     131      154     drew
18    1742      142     garyvee
19     905      136     buzzedition
20    4009      132     iconic88
21    1368      127     foursquare
22    1378      122     billzucker
23     840      121     heykim
24    4864      116     raybeckerman
25    2187      113     mayhemstudios
26     725      112     znatrainer
27    2507      112     evan
28    1241      110     crystal
29    2408      108     dudeman718
30     594      107     tweetmeme
31    2715      105     th
32    2658      101     tiger
33    1815       99     spam
34    2941       98     ded
35    3510       97     drewfromtv
36    1386       94     lukester
37    2604       94     al
38     460       92     americanwomannn
39     726       92     stop
40    4696       92     nytimes

and here the network graph:

From Drop Box

it is obvious that the one all neighbourhood of the twitter team cobntains a lot of other important (non twitter-team list) accounts that are partly more important than members of the twitter list themselves. Althoug derived from the twitter team list, the twitter team doesn’t play a main part but only a peripheral role. (they do the work others tweet about it
unfortunately I cannot extract so far a network with members of the twitter list only.

stop, there’s a work around:
I exctract the surrounding network, but don’t inlcude messages between neighbours
by unchecking “Connect Nodes”

——————————–
1 1771 1011 ev
2 3771 861 biz
3 2468 291 delbius
4 833 289 al3x
5 2046 245 caroline
6 1273 207 nk
7 4740 199 charles
8 944 196 jess
9 1341 180 laura
10 466 156 twitterapi
11 2510 112 evan
12 1235 110 crystal
13 2714 105 th
14 2231 101 tiger
15 2943 98 ded
16 1385 94 lukester
17 2607 94 al
18 4347 92 stop
19 1818 90 rael
20 4989 83 vl
21 5141 82 kevinweil
22 3862 79 bs
23 3322 76 goldman
24 4681 72 rsarver
25 1750 63 magnuson
26 3271 62 joshelman
27 3044 61 dougw
28 2110 57 sean
29 5161 55 trammell
30 1923 54 sam
31 1739 50 stevej
32 4646 50 ablegrape
33 4234 50 noradio
34 2450 49 lauraigomez
35 4243 49 che
36 4544 44 kevinthau
37 2587 43 edgutman
38 1925 42 anm
39 4102 42 mutgoff
40 237 39 netik

all top 40 are accounts from http://twitter.com/twitter/team

here’s how this network looks like:

From Drop Box

May 2010: now it is very easy to follow conversations among the twitter list members our brand new tool:

http://tlx.mememapper.com/#twitter/team

for more info how this analysis was done go to:
Community Core Detection in Twitter - a “Bottom Up” Heuristic

available at http://code.google.com/apis/chart/

Today i travel to Brussls and I’ll be [there]

Einladung zur zweiten deutschen Konferenz für Informationsarchitektur, die am 9. und 10. November 2007 unter dem Titel „Information Raum geben“ bei der Hochschule der Medien, Stuttgart (HdM) zu Gast ist. Die diesjährige Konferenz bildet gleichzeitig das Programm des 6. Symposiums für Informationsdesign und wird gemeinsam vom Studiengang Informationsdesign an der HdM und dem Institut für Informationsarchitektur veranstaltet.

mehr infos [hier]

Last Thursday we have been guests of Gerhard Dirmoser, who showed us his impressing collection of diagrams and his diagrammatic library. Gerhard is one of the leading experts in the field of diagrammatic and is devoting his work to the development of a new epistemological approach to describe and order diagrams. This approach is outstanding, because it aims to work finally without textual description, only on diagrammatic relations. Therefore probably the word “description” is inappropriate at all, because in Gerhards studio you realize that his research process consists in ordering, relating and placing objects, very similar to Aby Warburgs Mnemosyne. (see also german wikepedia entry on Warburg)
Aby Warburg revolutionised art history by introducing replications for didactic purposes. Nowadays image processing and graph engines can produce new experiences of exploring art. Gerhard Dirmosers and Dietmar Offenhuber project SemaSpace is exactly about the question of exploring semantically structured data and memory spaces. Dietmar Offenhuber convincingly solved the problem of handling large amounts of nodes, even several thousands – and even if the nodes are represented by images. Here’s a short description of SemaSpace by the authors:

SemaSpace is a fast and easy to use graph editor for large knowledge networks, specially designed for the application in non technical sciences and the arts. It creates interactive graph layouts in 2d and 3d by means of a flexible algorithm. The system is powerful enough for the calculation of complex networks and can incorporate additional data such as images, sounds and full texts.

On the SemaSpace Website you will find not only the tool but also an interesting application:

“25 years of ars electronica
study conducted by Gerhard Dirmoser, contains all projects / people involved in ars electronica until 2003, based on collected material and data from the ars electronica database. original files of the study:”

But SemaSpace is more than an organised database. It represents a “space of memory” that commemorates the threads of theory and media art within the “ars electronica universum.” It can be seen in the tradition Giulio Camillos Memory Theatre (see also http://www.clausmoser.com/?p=378) (By the way Camillo is a must for interaction designers)

Dietmar is currently working on a new version of SemaSpace and Gerhard is now about to network his collection of 4000 diagrams within the graph editor. As already two thirds of the work has been done within 20 workdays it is quite obvious that it seems an appropriate way to organise large amount of image data in a reasonable time span.

There’s a lot of other work (texts, diagrams and network graphs) by Gerhard available here: http://www.servus.at/kontext/ARS/ (strongly recommended).

Special hint for us lucky Austrians: next Sunday, February 4, a whole day lecture takes place at Audi Max of Danube University Krems.

Here are some remarks on BuzzFeed after having tested it for some days.
First of all, it does what it promises: It feeds you with buzz.
The term buzz itself implies that there is a greater audience behind. It will not easily become a buzz when two mathematicians are discussing problems in algebra. Buzz needs a bigger number of people that talk about it and a potential to infect even more people. A buzz from the last year is no longer a buzz. BuzzFeed detects buzz before it becomes a bigger thing.

BuzzFeed is showing buzz a few moments before its tipping point. From an analytical perspective it is already clear that a new buzz is emerging, but the masses don’t know it yet. BuzzFeed is therefore an adequate means of keeping up with the public opinion and to be some eye glimpses ahead. It’s an accelerator of public discourse. But do we really need even more accelerators? Tools like BuzzFeed make it very clear that the blogosphere is a huge discourse machine and that its speed and effectiveness is growing. The whole machinery is based on the simple fact that communication is producing communication; sometimes a cascade of communication; But what is the outcome of it? Doesn’t that lead to a more and more superficial mode of communication? Does it make us more fit to face the challenges of a crazy and complex world or is it just another step to make it a bit more crazy and complex? To be honest I don’t know.

Under the bottom line – and beyond all sociological considerations (sorry I couldn’t withstand) - BuzzFeed is simply an vanguard media. It typically combines the following components:
1. Consumer Generated Media - mainly weblogs – that provide ever new content; in this case CGM is working like an armada of journalists chasing for latest news. Or to put it in other words: they work like sensor neurons in our nervous system that fire when they perceive a stimulus.
2. Analytical tools detect trends within the blogosphere. In the case of BuzzFeed these tools detect upcoming topics as patterns. Patterns mean that there’s not a chaotic sequence of “firing neurons” but there’s something going on; something that needs further interpretation.
3. Obviously these patterns are not self-explaining and require some training to interpret them. Therefore BuzzFeed hire editors to separate the wheat from the chaff and to write short introductory texts to topics they consider being upcoming and interesting enough to get featured.

BuzzFeed is therefore a hybrid media that combines a very large network of writers, computational power and human judgement. The latter seems not to be replaceable by technology and is still the key factor that makes a project juicy. We may expect many more interesting combinations of theses three components that make up new web media formats, not only including text but also podcasts and video.

An cool way to explore html-code is the htmlgraph (Websites as Graph).
It is written in processing and traer.physics which is a particle system physics engine for processing.