Semi-Supervised SVM With Extended Hidden Features

Many traditional semi-supervised learning algorithms not only train on the labeled samples but also incorporate the unlabeled samples in the training sets through an automated labeling process such as manifold preserving. If some labeled samples are falsely labeled, the automated labeling process wi...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE transactions on cybernetics 2016-12, Vol.46 (12), p.2924-2937
Hauptverfasser:	Dong, Aimei, Chung, Fu-Lai, Deng, Zhaohong, Wang, Shitong
Format:	Artikel
Sprache:	eng
Schlagworte:	Algorithms Automation Convergence Cybernetics Extremely high frequencies Hidden features Hidden object recognition integrated squared error between probability distributions Labeling Labelling Linear programming Machine learning maximum margin Probability distribution semi-supervised learning (SSL) support vector machine (SVM) Support vector machines Training
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!