In this work we propose a novel parametric Bayesian model to solve the problem of semi-supervised learning, including classification and clustering. Standard ap- proaches of semi-supervised classification can recognize classes but cannot find groups of data. On the other hand, semi-supervised clustering techniques are able to discover groups of data but cannot find the associations between clusters and classes. The proposed model can classify and cluster samples simultaneously, leading to the possibility of solving the problem of annotation in the presence of an unknown number of classes and/or arbitrary number of clusters per class. Preliminary results performed on synthetic datasets show the effectiveness of the framework.

Classtering: A Semi-Supervised Algorithm Based on Mixture of Factor Analysers

Sansone, Emanuele;De Natale, Francesco
2015

Abstract

In this work we propose a novel parametric Bayesian model to solve the problem of semi-supervised learning, including classification and clustering. Standard ap- proaches of semi-supervised classification can recognize classes but cannot find groups of data. On the other hand, semi-supervised clustering techniques are able to discover groups of data but cannot find the associations between clusters and classes. The proposed model can classify and cluster samples simultaneously, leading to the possibility of solving the problem of annotation in the presence of an unknown number of classes and/or arbitrary number of clusters per class. Preliminary results performed on synthetic datasets show the effectiveness of the framework.
Trento
DISI
Sansone, Emanuele; De Natale, Francesco
File in questo prodotto:
File Dimensione Formato  
Classtering.pdf

accesso aperto

Descrizione: Articolo principale
Tipologia: Versione editoriale (Publisher’s layout)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 1.39 MB
Formato Adobe PDF
1.39 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11572/106693
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact