Semi-Supervised Learning of Mixture Models

Fabio Gagliardi Cozman, Ira Cohen, Marcelo Cesar Cirelo

Resultado de la investigación: Contribución a una conferenciaArtículo de conferencia

139 Citas (Scopus)

Resumen

This paper analyzes the performance of semi-supervised learning of mixture models. We show that unlabeled data can lead to an increase in classification error even in situations where additional labeled data would decrease classification error. We present a mathematical analysis of this "degradation" phenomenon and show that it is due to the fact that bias may be adversely affected by unlabeled data. We discuss the impact of these theoretical results to practical situations.
Idioma originalInglés estadounidense
Páginas99-106
Número de páginas8
EstadoPublicada - 1 dic 2003
Publicado de forma externa
EventoProceedings, Twentieth International Conference on Machine Learning -
Duración: 1 dic 2003 → …

Conferencia

ConferenciaProceedings, Twentieth International Conference on Machine Learning
Período1/12/03 → …

Huella Profundice en los temas de investigación de 'Semi-Supervised Learning of Mixture Models'. En conjunto forman una huella única.

  • Citar esto

    Cozman, F. G., Cohen, I., & Cirelo, M. C. (2003). Semi-Supervised Learning of Mixture Models. 99-106. Papel presentado en Proceedings, Twentieth International Conference on Machine Learning, .