Please use this identifier to cite or link to this item: http://repositorio.unicamp.br/jspui/handle/REPOSIP/326324
Type: Artigo
Title: Improving Semi-supervised Learning Through Optimum Connectivity
Improving semi-supervised learning through optimum connectivity
Author: Amorim, Willian P.
Falcão, Alexandre X.
Papa, João P.
Carvalho, Marcelo H.
Abstract: The annotation of large data sets by a classifier is a problem whose challenge increases as the number of labeled samples used to train the classifier reduces in comparison to the number of unlabeled samples. In this context, semi-supervised learning methods aim at discovering and labeling informative samples among the unlabeled ones, such that their addition to the correct class in the training set can improve classification performance. We present a semi-supervised learning approach that connects unlabeled and labeled samples as nodes of a minimum-spanning tree and partitions the tree into an optimum-path forest rooted at the labeled nodes. It is suitable when most samples from a same class are more closely connected through sequences of nearby samples than samples from distinct classes, which is usually the case in data sets with a reasonable relation between number of samples and feature space dimension. The proposed solution is validated by using several data sets and state-of-the-art methods as baselines. (C) 2016 Elsevier Ltd. All rights reserved.
The annotation of large data sets by a classifier is a problem whose challenge increases as the number of labeled samples used to train the classifier reduces in comparison to the number of unlabeled samples. In this context, semi-supervised learning meth
Subject: Processamento de imagens
Inteligência artificial
Floresta de caminhos ótimos
Aprendizado de máquina
Reconhecimento de padrões
Country: Países baixos
Editor: Elsevier
Citation: Pattern Recognition. Elsevier Sci Ltd, v. 60, p. 72 - 85, 2016.
Rights: fechado
Fechado
Identifier DOI: 10.1016/j.patcog.2016.04.020
Address: https://www.sciencedirect.com/science/article/pii/S0031320316300668
Date Issue: 2016
Appears in Collections:IC - Artigos e Outros Documentos

Files in This Item:
File SizeFormat 
000383525600008.pdf3.29 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.