MinerLC: closed pattern miner in attributed graphs
Dominique Bouthinon, Guillaume Santini, Henry Soldano
This archive contains a Makefile and a directory with a french documentation that describes the formats of input and output files and explains how to compile and run the minerLC and associated programs like supportsAttendus.
- Social network from : The Teenage Friends and Lifestyle Study (Bush, West & Michell 1997, Michell and West 1996, Pearson and Michell 2000, Pearson and West 2003) https://www.stats.ox.ac.uk/~snijders/siena/Glasgow_data.htm
- DBLP co-authorship and LastFM from : Esther Galbrun, Aristides Gionis, and Nikolaj Tatti (2014) Overlapping community detection in labeled graphs.
- DBLP co-authorship labelled graph inspired from : Adriana Prado, Marc Plantevit, Celine Robardet, Jean-Francois Boulicaut (2012) Mining Graph Topological Patterns: Finding Co-Variations Among Vertex Descriptors
- Labelled graphs from : Arlei Silva, Wagner Meira, Mohammed J. Zaki (2010) Structural correlation pattern mining for large graphs (http://www.cs.ucsb.edu/~arlei/)
- Corporate law partnership advice network.
- Supreme court precedent network
- Citation networks for satisticians :
Command line examples
A command line to compute the 4-4 Hub Authority core closed patterns. The program output is redirected from the standard output to a results file lawer_advice.o_4_4
minerLC lawer_advice.nri -o -hub_aut1 4 4 > lawer_advice.o_4_4
A command line to compute the expected size of the abstract closed pattterns previously computed from 1000 random samples. The program output is redirected from the standard output to a results file lawer_advice.o_4_4_1000
supportsAttendus lawer_advice.nri lawer_advice.o_4_4 -o -hub_aut1 4 4 -n 1000 > lawer_advice.o_4_4_1000
CDF wolfram computable document (interactive)
Hereunder two CDF documents allowing to investigate interactively the 4-4 and 9-9 Hub Authority core closed patterns.
In order to "play" the interactive documents you need to download and install the wolfram CDF player from the following URL:
In case of difficulties please contact us at: santini -at- lipn.univ-paris13.fr