Article


Article Code : 1395061719063321(DOI : 10.7508/jist.2017.18.005)

Article Title : Node Classification in Social Network by Distributed Learning Automata

Journal Number : 18 Spring 2017

Visited : 1208

Files : 968 KB


List of Authors

  Full Name Email Grade Degree Corresponding Author
1 Ahmad Rahnama Zadeh rahnamazade@gmail.com Post Graduate Student M.Sc
2 Mohammad Reza Meybodi mmeybodi@aut.ac.ir Associate Professor PhD
3 Masoud Taheri Kadkhoda taheri.masood@gmail.com Graduate M.Sc

Abstract

The aim of this article is improving the accuracy of node classification in social network using Distributed Learning Automata (DLA). In the proposed algorithm using a local similarity measure, new relations between nodes are created, then the supposed graph is partitioned according to the labeled nodes and a network of Distributed Learning Automata is corresponded on each partition. In each partition the maximal spanning tree is determined using DLA. Finally nodes are labeled according to the rewards of DLA. We have tested this algorithm on three real social network datasets, and results show that the expected accuracy of presented algorithm is achieved.