®®®® SIIA Público

Título del libro: Iisa 2015 - 6th International Conference On Information, Intelligence, Systems And Applications
Título del capítulo: Why the Naive Bayes approximation is not as Naive as it appears

Autores UNAM:
CHRISTOPHER RHODES STEPHENS;
Autores externos:

Idioma:
Inglés
Año de publicación:
2016
Palabras clave:

Artificial intelligence; Data mining; Errors; Learning systems; Attribute sets; Classifier performance; Error measures; Likelihood functions; Local error; Naive bayes; Problem domain; Robust performance; Classifiers


Resumen:

The Naive Bayes approximation and associated classifier is widely used in machine learning and data mining and offers very robust performance across a large spectrum of problem domains. As it depends on a very strong assumption - independence among features - this has been somewhat puzzling. Various hypotheses have been put forward to explain its success and moreover many generalizations have been proposed. In this paper we propose a set of "local" error measures - associated with the likelihood functions for particular subsets of attributes and for each class - and show explicitly how these local errors combine to give a "global" error associated to the full attribute set. By so doing we formulate a framework within which the phenomenon of error cancelation, or augmentation, can be quantitatively evaluated and its impact on classifier performance estimated and predicted a priori. These diagnostics also allow us to develop a deeper and more quantitative understanding of why the Naive Bayes approximation is so robust and under what circumstances one expects it to break down. © 2015 IEEE.


Entidades citadas de la UNAM: