Efficient Mining of Subsample-Stable Graph Patterns

Aleksey Buzmakov 1 Sergei Kuznetsov 2 Amedeo Napoli 3
3 ORPAILLEUR - Knowledge representation, reasonning
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : A scalable method for mining graph patterns stable under subsampling is proposed. The existing subsample stability and robustness measures are not antimonotonic according to definitions known so far. We study a broader notion of anti-monotonicity for graph patterns, so that measures of subsample stability become antimonotonic. Then we propose gSOFIA for mining the most subsample-stable graph patterns. The experiments on numerous graph datasets show that gSOFIA is very efficient for discovering subsample-stable graph patterns.
Document type :
Conference papers
Complete list of metadatas

Cited literature [22 references]  Display  Hide  Download

https://hal.inria.fr/hal-01668663
Contributor : Aleksey Buzmakov <>
Submitted on : Wednesday, December 20, 2017 - 11:00:13 AM
Last modification on : Tuesday, December 18, 2018 - 4:38:02 PM

File

icdm17-Sofia-graph.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01668663, version 1

Collections

Citation

Aleksey Buzmakov, Sergei Kuznetsov, Amedeo Napoli. Efficient Mining of Subsample-Stable Graph Patterns. ICDM 2017 - 17th IEEE International Conference on Data Mining, Nov 2017, New Orleans, United States. pp.1-6. ⟨hal-01668663⟩

Share

Metrics

Record views

297

Files downloads

239