Busca avançada
Ano de início
Entree


A Comparison of Graph Construction Methods for Semi-Supervised Learning

Texto completo
Autor(es):
Berton, Lilian ; Lopes, Alneu de Andrade ; Vega-Oliveros, Didier A. ; IEEE
Número total de Autores: 4
Tipo de documento: Artigo Científico
Fonte: 2018 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN); v. N/A, p. 8-pg., 2018-01-01.
Resumo

Graph-based methods are among the most active approaches to semi-supervised learning. This occurs mainly due to their ability to deal with local and global characteristics of available data, identify classes or groups regardless the data shape, and to represent submanifold in Euclidean space. Graph-based methods are sensitive to graph construction and a challenge in the area is the construction of a graph to represent data patterns. Several unsupervised graph construction methods have been proposed for dealing with different issues. However, it lacks a detailed study that evaluates their properties and effectiveness. Here, we analyze the robustness of such methods for SSL classification. The graph construction methods analyzed include k-nearest neighbor (kNN), mutual kNN combined with minimum spanning tree (M-kNN), b-matching by belief propagation (BP), b-matching by greedy approximation and sequential kNN (S-kNN). Statistical analyses are carried out with respect to classification accuracy. We observe that robustness of the methods varies according to external factors, such as labeled data representativeness and parameters k or b, and internal factors, related to the topology of the network. Regular graph construction methods, b-matching and S-kNN, achieve the best results on classification and generate more homogeneous and sparse networks. (AU)

Processo FAPESP: 16/23698-1 - Processos Dinâmicos em Aprendizado de Máquina baseados em Redes Complexas
Beneficiário:Didier Augusto Vega Oliveros
Modalidade de apoio: Bolsas no Brasil - Pós-Doutorado
Processo FAPESP: 15/14228-9 - Análise e Mineração de Redes Sociais
Beneficiário:Alneu de Andrade Lopes
Modalidade de apoio: Auxílio à Pesquisa - Regular
Processo FAPESP: 11/21880-3 - Construção de redes para o aprendizado semissupervisionado
Beneficiário:Lilian Berton
Modalidade de apoio: Bolsas no Brasil - Doutorado