Switch branches/tags
Nothing to show
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
README
report.txt
survey_numeric_before_edge_list.txt
survey_numeric_combined_edge_list.txt
survey_numeric_during_edge_list.txt
survey_numeric_status_map.csv

README

Dataset:
   A network derived from a survey taken at a doctoral summer school.
   The data was gathered and prepared by AaronMcDaid, Samantha Lam,
   Ursula Redmond and Arthur White[1].

Current Version:
   2012-03-28 First public release

Webpage:
   http://sites.google.com/site/aaronmcdaid/sbm
   http://clique.ucd.ie/data (along with other datasets from our research group)

Created by:
   Aaron McDaid, Samantha Lam, Ursula Redmond, Arthur White.
   Clique Research Cluster, UCD Dublin and DERI Galway, Ireland
   This research was supported by Science Foundation Ireland (SFI) Grant No. 08/SRC/I1407 - Clique Research Cluster


== Description ==

The 74 attendees were asked who they knew before the summer school and were
also asked who they met during the summer school. 40 responded and gave us
permission to make their data available in this anonymized dataset.
We have completely removed one person who responded but did not give such
permission.  This gives us a network of 73 nodes, where node A has a link
to node B if person A reported that they knew (or met) person B.

There are three directed, unweighted, networks in this first release:

survey_numeric_before_edge_list.txt:
   source node knew target node before the summer school.
    583 edges between 67 nodes
survey_numeric_during_edge_list.txt:
   source node met target node at the summer school.
    912 edges between 71 nodes
survey_numeric_combined_edge_list.txt:
   the union of the two networks above
   1138 edges between 73 nodes

There are 40 nodes with a non-zero out-degree in the combined network, these
are the 40 that responded and gave us permission.

Each of the 73 nodes has non-zero degree in the combined network,
but some nodes are effetively missing from the other networks as they
had zero degree in those networks.

The files are a simple edge list, where each node has been assigned
a number between 0 and 72 inclusive. We also did some annotation
of the formal role of the person at the summer school.

survey_numeric_status_map.csv:
   This records, for each of the 73 people ordered from 0 to 72, what
   their primary role was:
      Speaker: Presented at the summer school
      PC: Programme Committee
      LO: Local organizing commitee
      Attendee: Didn't fit into the above three roles.
   There is also a field recording secondary_roles for those people who
   fitted into multiple roles.

== Analysis ==

This dataset is analyzed and presented in two papers, which are under
review as of 2012-04-02:

 -  "Clustering in networks with the collapsed Stochastic Block Model"
This describes a clustering algorithm which clustered the nodes based
on the network. The clusters found were found to approximately match
the roles described in survey_numeric_status_map.csv.
 -  "Increasing the Scalability of the Fitting of Generalised Block Models for Social Networks"

Bibtex:
@article{CollapsedSBM,
    archivePrefix = {arXiv},
    author = {McDaid, Aaron F. and Murphy, Brendan and Friel, Nial and Hurley, Neil},
    title = {Clustering in networks with the collapsed Stochastic Block Model},
    url = {http://arxiv.org/abs/1203.3083},
    day = {14}, eprint = {1203.3083}, month = mar, posted-at = {2012-03-15 10:10:36}, priority = {2},
    year = {2012}
}
(This paper is under review as of 2012-04-02.)
@inproceedings{ChanGBM,
    author = {Chan, Jeffrey and Lam, Samantha and Hayes, Conor},
    booktitle = {International Joint Conference on Artificial Intelligence 2011 (IJCAI)},
    priority = {2},
    title = {Increasing the Scalability of the Fitting of Generalised Block Models for Social Networks},
    url = {http://www.deri.ie/fileadmin/documents/uimr/jkcchan\_icjai11.pdf},
    year = {2011}
}
(A later version of this paper is under review as of 2012-04-02.)




References:

[1] AaronMcDaid <aaronmcdaid@gmail.com>, Arthur White <arthurwhite87@hotmail.co.uk>,
        Ursula Redmond <u.redmond@gmail.com>, Samantha Lam <contradirony@gmail.com>
[2] http://arxiv.org/abs/1203.3083
[3] https://sites.google.com/site/aaronmcdaid/sbm
[4] http://clique.ucd.ie/data    Includes other datasets from our research group.