User Tools

Site Tools


Sidebar

Practical Information:

Teaching:

Bâtiment Nautibus
43, Bd du 11 Novembre 1918
69622 Villeurbanne Cedex.
☏: +33(0)472 43 16 35
email: marc.plantevit-at-univ-lyon1.fr

Research:

Bureau 501.319
Bâtiment Blaise Pascal
7, Avenue Jean Capelle
69621 Villeurbanne Cedex
☏: +33(0)472 43 84 87
Fax: +33(0)472 43 87 13
email: marc.plantevit-at-liris.cnrs.fr

data_sets

Static Attributed Graphs

Co-authorship network

This data set depicts a co-authorship graph built from the DBLP digital library. Each vertex represents an author who published at least one paper in one of the major conferences and journals of the Data Mining and Database communities between January 1990 and February 2011. Each edge links two authors who co-authored at least one paper (no matter the conference or journal). The vertex properties are the number of publications in each of the 29 selected conferences or journals and we also consider 9 topological properties:

Attributes
Conferences JournalsTopological Properties
KDD, ICDM, ECML/PKDD, PAKDD, SIAM DM, AAAI, ICML,IJCAI, IDA, DASFAA, VLDB, CIKM, SIGMOD, PODS, ICDE, EDBT, ICDT, SAC IEEE TKDE, DAMI, IEEE Int. Sys., SIGKDD Exp., Comm. ACM,IDA J., KAIS, SADM, PVLDB, VLDB J., ACM TKDD Degree Cent., Closeness Cent., Betweenness Cent., EigenVector Cent., PageRank, Clustering Coeff., Size of Max. Quasi-Clique, Number of Quasi-Cliques, Size of Community

This attributed graph contains 42,252 vertices and 210,320 undirected edges.

If you are interested to use this data set, send me an email and I will give you a link to download the data.

Please cite this reference if you use this data set:

<note>Adriana Prado, Marc Plantevit, Celine Robardet, Jean-Francois Boulicaut, “Mining Graph Topological Patterns: Finding Co-Variations Among Vertex Descriptors,”. IEEE Transactions on Knowledge and Data Engineering, 06 Aug. 2012. IEEE computer Society Digital Library. IEEE Computer Society, <http://doi.ieeecomputersociety.org/10.1109/TKDE.2012.154> </note>

Dynamic Attributed Graphs

Dynamic Co-authorship network

This dataset is a subset of the DBLP dataset (http://dblp.uni-trier.de/). Vertices of the graph are authors and an edge exists between them if the corresponding authors have written a paper together in a given period of time. Only authors who had at least 10 publications (in a selected set of 43 conferences/journals) from 1990 to 2010 are considered. There are in total 2,723 authors. Each graph depicts co-authorship relations over 5 years ([1990-1994][1992-1996][1994-1998][1996-2000][1998-2002][2000-2004][2002-2006][2004-2008][2006-2010]). Notice that we consider overlapping time periods to maintain a coherence in the co-authorship relations. Each vertex at each time is associated to a set of 43 attributes corresponding to the number of publications in each conference/journal during the related period. To summarize, this dataset consists in 2,723 vertices, 9 timestamps and 43 attributes.

If you are interested to use this data set, send me an email and I will give you a link to download the data.

Please cite this reference if you use this data set: <note>Elise Desmier, Marc Plantevit, Céline Robardet and Jean-François Boulicaut. Cohesive Co-Evolution Patterns in Dynamic Attributed Graphs. In Proceedings of Discovery Science (DS'12), 2012. </note>

data_sets.txt · Last modified: 2013/10/15 07:03 by mplantev

CNRS INSA de Lyon Université Lyon 1 Université Lyon 2 École centrale de Lyon