We present a novel approach for the detection of 2-plexes, a popular relaxation of cliques used for modeling network communities. Specifically, with the purpose of identifying theoretically sound methods for community detection on a large scale, we introduce the first shared-nothing distributed algorithm for this problem. This result opens a new research direction for scalable community detection. Our proposal has three main ingredients: (i) we reduce the problem of finding 2-plexes to that of finding cliques; (ii) we leverage known algorithms for fast computation of cliques; (iii) we exploit a decomposition technique for a distributed shared-nothing computation. Preliminary experiments on a 10-nodes cluster running Spark confirm the effectiveness of our approach.
Conte, A., Patrignani, M., Firmani, D., Torlone, R. (2019). Shared-nothing distributed enumeration of 2-plexes. In 28th ACM International Conference on Information and Knowledge Management (CIKM), Class A+ (GII-GRIN rating) (pp.2469-2472). Association for Computing Machinery [10.1145/3357384.3358083].
Shared-nothing distributed enumeration of 2-plexes
Conte Alessio;Patrignani Maurizio;Firmani Donatella;Torlone Riccardo
2019-01-01
Abstract
We present a novel approach for the detection of 2-plexes, a popular relaxation of cliques used for modeling network communities. Specifically, with the purpose of identifying theoretically sound methods for community detection on a large scale, we introduce the first shared-nothing distributed algorithm for this problem. This result opens a new research direction for scalable community detection. Our proposal has three main ingredients: (i) we reduce the problem of finding 2-plexes to that of finding cliques; (ii) we leverage known algorithms for fast computation of cliques; (iii) we exploit a decomposition technique for a distributed shared-nothing computation. Preliminary experiments on a 10-nodes cluster running Spark confirm the effectiveness of our approach.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.