We propose a Support Vector-based methodology for learn- ing classifiers from partially labeled data. Its novelty stands in a formulation not based on the cluster hypothesis, stating that learning algorithms should search among classifiers whose decision surface is far from the unlabeled points. On the contrary, we assume such points as specimens of uncertain labels which should lay in a region containing the decision surface. The proposed approach is tested against synthetic data sets and subsequently applied to well-known benchmarks, attaining better or at least comparable performance w.r.t. methods described in the literature.
Avoiding the cluster hypothesis in SV Classification of partially labeled data / D. Malchiodi, T. Legnani (SMART INNOVATION, SYSTEMS AND TECHNOLOGIES). - In: Recent advances of neural network models and applications : proceedings of the 23rd Workshop of the Italian Neural Networks Society (SIREN), May 23 - 25, Vietri sul Mare, Salerno, Italy / [a cura di] S. Bassis, A. Esposito, F.C. Morabito. - Heidelberg : Springer, 2014. - ISBN 9783319041285. - pp. 33-40 (( Intervento presentato al 23. convegno Workshop of the Italian Neural Networks Society (SIREN) tenutosi a Vietri sul Mare nel 2013 [10.1007/978-3-319-04129-2_4].
Avoiding the cluster hypothesis in SV Classification of partially labeled data
D. Malchiodi
;
2014
Abstract
We propose a Support Vector-based methodology for learn- ing classifiers from partially labeled data. Its novelty stands in a formulation not based on the cluster hypothesis, stating that learning algorithms should search among classifiers whose decision surface is far from the unlabeled points. On the contrary, we assume such points as specimens of uncertain labels which should lay in a region containing the decision surface. The proposed approach is tested against synthetic data sets and subsequently applied to well-known benchmarks, attaining better or at least comparable performance w.r.t. methods described in the literature.File | Dimensione | Formato | |
---|---|---|---|
shadowed.pdf
accesso riservato
Descrizione: Articolo
Tipologia:
Publisher's version/PDF
Dimensione
242.97 kB
Formato
Adobe PDF
|
242.97 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.