Preventing error propagation in semi-supervised learning

Authors:
Thiago C. Silva;Liang Zhao
Affiliations:
Department of Computer Science, Institute of Mathematics and Computer Science, University of Säo Paulo, Säo Carlos, SP, Brazil;Department of Computer Science, Institute of Mathematics and Computer Science, University of Säo Paulo, Säo Carlos, SP, Brazil
Venue:
ISNN'12 Proceedings of the 9th international conference on Advances in Neural Networks - Volume Part I
Year:
2012

Citing 5
Cited 0

Semi-supervised learning with an imperfect supervisor

Knowledge and Information Systems
Semi-Supervised Learning (Adaptive Computation and Machine Learning)

Semi-Supervised Learning (Adaptive Computation and Machine Learning)
Learning from imperfect data

Applied Soft Computing
Label Propagation through Linear Neighborhoods

IEEE Transactions on Knowledge and Data Engineering
Semi-supervised learning guided by the modularity measure in complex networks

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Semi-supervised learning is a machine learning approach that is able to employ both labeled and unlabeled samples in the training process. In these cases, the reliability of the labels is a crucial factor, because mislabeled samples may propagate wrong labels to a portion of or even the entire data set. This paper has the objective of addressing the error propagation problem originated by these mislabeled samples by presenting a mechanism embedded in a network-based (graph-based) semi-supervised learning method. Such a technique is based on a combined random-preferential walk of particles in a network constructed from the input data set. The particles of the same class cooperate among them, while the particles of different classes compete with each other to propagate class labels to the whole network. Computer simulations conducted on real-world data sets reveal the effectiveness of the model.