Author Archives: jasonjjones

What Does a Political Networks Conference Do? (On Twitter)

Last week I mapped the Twitter follow network of those attending the Political Networks Conference in Boulder, CO.  I gathered this data prior to the first day of the conference, so I decided to take a second look now that the conference is over.

Political Networks Twitter Follow Network – Post Conference (Click image for larger version.)

How does the post-conference graph differ from the pre-conference graph?

  • The first notable difference is the addition of 18 new nodes.  The size of the network grew by over 50% as new Twitter users decided to follow @PolNetworks during the conference.
  • Another difference is the addition of 215 new follow edges.  With only 181 edges in the original graph, the number of follow relationships in the network grew more than 100%.
  • The account with the biggest increase in followers was @krmckelv.  Karissa went from only 1 follower to 8.  (She has since changed her Twitter handle to @karissamckelvey.)

Now let’s limit the post-conference graph to a subgraph containing only those nodes that were also in the pre-conference graph.  How has this component changed?

Graph Density: Increased 0.03 from 0.15 to 0.18.

Graph Transitivity:  Decreased 0.03 from 0.56 to 0.53.

Graph Efficiency:  Decreased 0.02 from 0.87 to 0.85.

So new follow edges were added (density), but in a way that more incomplete than complete triangles were formed (transitivity).  Not surprisingly, the new ties were redundant in connecting the graph, lowering efficiency.

Here are some individual-level statistics for the interested.  It was great to get to know many of you at the conference, and I’ll see you at the next one!

Account In Degree Out Degree Eig. Centrality
JaciKettler 10 20 0.3
smotus 35 18 0.27
kwcollins 20 19 0.27
burtmonroe 15 17 0.26
therriaultphd 20 18 0.24
BrendanNyhan 31 17 0.24
JohnCluverius 8 13 0.22
RebeccaHannagan 4 12 0.21
ianpcook 7 12 0.2
Student 27 14 0.2
jlove1982 7 11 0.2
KyleLSaunders 9 13 0.19
krmckelv 12 17 0.19
JeffGulati 9 12 0.18
richardmskinner 8 10 0.15
James_H_Fowler 8 11 0.15
prisonrodeo 16 9 0.15
archimedino 5 10 0.14
davekarpf 7 8 0.13
NateMJensen 5 8 0.13
jon_m_rob 0 8 0.13
sissenberg 8 7 0.12
3876 7 8 0.12
hsquared47 3 7 0.12
marioguerrero 1 7 0.12
FHQ 17 7 0.11
heathbrown 4 8 0.11
First_Street 1 7 0.1
98percentright 5 6 0.1
davidlazer 17 5 0.07
GeoffLorenz 0 5 0.07
jasonjones_jjj 4 6 0.06
matthewhitt 2 4 0.06
DocPolitics 1 4 0.06
JoeLenski 2 4 0.05
maizeandblue 1 4 0.05
allenlinton2 2 3 0.04
DryHeathen 0 4 0.04
ajungherr 0 4 0.04
slimbock 0 2 0.03
cassyld 2 2 0.02
PolNetworks 52 1 0.02
rmbond15 2 3 0.02
DominikBatorski 0 2 0.01
vatrafilm 0 1 0
janschulz 0 1 0
dogaker 0 1 0
stefanjwojcik 1 1 0
KenneyMkenney 0 1 0
ophastings 0 1 0
jbrittaq 0 1 0
jboxstef 1 1 0
Hirschi5 0 1 0

Download edge list as .xlsx: polnetworks2_edge_list


Acquisition of Social Network Structure

by Jason J. Jones
Scale Free Graph

On Tuesday at the HNG meeting I’ll be discussing three experiments I’ve conducted on the acquisition of social network structure. For a preview, you can read my submission to CogSci 2011 which discusses one of the experiments.

Congressional Speech Corpus (including references to other members of Congress)

By Jason J. Jones

I ran across this corpus of Congressional speech that may be useful to some in the group.  Here is a brief description:

This data includes speeches as individual documents, together with:

  • automatically-derived labels for whether the speaker supported or opposed the legislation discussed in the debate the speech appears in, allowing for experiments with this kind of sentiment analysis
  • indications of which “debate” each speech comes from, allowing for consideration of conversational structure
  • indications of by-name references between speakers, and the scores that our agreement/disagreement classifier(s) automatically assigned to such references, allowing for experiments on agreement classification if one assigns “true” labels from the support/oppose labels assigned to the pair of speakers in question
  • the edge weights and other information we derived to create the graphs we used for our experiments upon this data, facilitating implementation of alternative graph-based methods upon the graphs we constructed

The third bullet seems like it would be of particular interest.

In my data mining class we are not using this corpus, unfortunately.  But, if you want to know which words most likely indicate an unfavorable movie review, I should have a classifier that will tell you by next week.

Reporting the News Using Only Twitter and Facebook as Sources

By Jason J. Jones

Five journalists have volunteered to be locked in a house with no access to the outside world except Facebook and Twitter. Their task will be to report the day’s news as best they can using only these sources.

The Washington Post and Ars Technica are optimistic.  Personally, I wouldn’t be surprised if the stories they report end up a little garbled.

SPSP Political Psychology Pre-Conference.

by Jason J. Jones

A one-day Political Psychology conference will take place in Las Vegas on the opening day of the Society for Personality and Social Psychology Annual Conference.

If anyone wants to go, I was planning on driving up there for SPSP anyway, and you are welcome to join me.  I even have an extra room in the timeshare I have for the weekend.  The pre-conference is on Thursday, Jan. 28th, and SPSP will continue on Friday and Saturday.  (Sunday we break the bank at the blackjack tables.)

DARPA Network Challenge

by Jason J. Jones

Wherever you are this Saturday – look up.  If you see a giant red orb, don’t be alarmed.  It just means you are part of the Defense Advanced Research Projects Agency‘s latest experiment.

The DARPA Network Challenge is

a competition that will explore the roles the Internet and social networking play in the timely communication, wide-area team-building, and urgent mobilization required to solve broad-scope, time-critical problems. The challenge is to be the first to submit the locations of 10 moored, 8-foot, red, weather balloons at 10 fixed locations in the continental United States.

A $40,000 prize will be awarded to whomever first submits the correct location (latitude and longitude) of all ten balloons.

Naturally, people are forming teams with varying strategies and varying success.  Many are offering bounties for each balloon location (e.g. Red Balloon Race).  Some are hoping people will participate just for the fun or challenge, and plan to donate the prize money to charity (e.g. Project Red Baloon).

Personally, I plan to sit at my computer in my pajamas all day Saturday and “watch” as it were.  I’ll be checking the Facebook groups and some of the bigger challenge-specific sites.   I may even have to join Twitter (shudder).

Keep watching the skies.  The balloons are out there…

A $40K DARPA Contest Involving Social Networks, Misinformation and Disinformation

“DARPA, in order to celebrate the 40th anniversary of the internet, has issued a zeitgeisty “Network Challenge”. The first person or team to correctly locate ten large red balloons scattered across the USA will win $40,000.”

Full story: