Semi-supervised learning guided by the modularity measure in complex networks


Autoria(s): Silva, Thiago Christiano; Liang, Zhao
Contribuinte(s)

UNIVERSIDADE DE SÃO PAULO

Data(s)

30/10/2013

30/10/2013

2012

Resumo

Semi-supervised learning techniques have gained increasing attention in the machine learning community, as a result of two main factors: (1) the available data is exponentially increasing; (2) the task of data labeling is cumbersome and expensive, involving human experts in the process. In this paper, we propose a network-based semi-supervised learning method inspired by the modularity greedy algorithm, which was originally applied for unsupervised learning. Changes have been made in the process of modularity maximization in a way to adapt the model to propagate labels throughout the network. Furthermore, a network reduction technique is introduced, as well as an extensive analysis of its impact on the network. Computer simulations are performed for artificial and real-world databases, providing a numerical quantitative basis for the performance of the proposed method.

State of Sao Paulo Research Foundation (FAPESP)

Brazilian National Council of Technological and Scientific Development (CNPq)

Identificador

NEUROCOMPUTING, AMSTERDAM, v. 78, n. 1, Special Issue, p. 30-37, FEB 15, 2012

0925-2312

http://www.producao.usp.br/handle/BDPI/36828

10.1016/j.neucom.2011.04.042

http://dx.doi.org/10.1016/j.neucom.2011.04.042

Idioma(s)

eng

Publicador

ELSEVIER SCIENCE BV

AMSTERDAM

Relação

NEUROCOMPUTING

Direitos

restrictedAccess

Copyright ELSEVIER SCIENCE BV

Palavras-Chave #SEMI-SUPERVISED LEARNING #MODULARITY #COMPLEX NETWORKS #NETWORK REDUCTION #COMMUNITY STRUCTURE #RANDOM-WALKS #COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE
Tipo

article

original article

publishedVersion