Back to Search
Start Over
When is the Naive Bayes approximation not so naive?
- Source :
- Machine Learning. 107:397-441
- Publication Year :
- 2017
- Publisher :
- Springer Science and Business Media LLC, 2017.
-
Abstract
- The Naive Bayes approximation (NBA) and associated classifier are widely used and offer 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 many generalizations have been proposed. In this paper we propose a set of "local" error measures--associated with the likelihood functions for 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 quantified and its impact on classifier performance estimated and predicted a priori. These diagnostics allow us to develop a deeper and more quantitative understanding of why the NBA is so robust and under what circumstances one expects it to break down. We show how these diagnostics can be used to select which features to combine and use them in a simple generalization of the NBA, applying the resulting classifier to a set of real world data sets.
- Subjects :
- business.industry
Generalization
02 engineering and technology
Bayes classifier
Machine learning
computer.software_genre
Naive Bayes classifier
Artificial Intelligence
020204 information systems
Phenomenon
0202 electrical engineering, electronic engineering, information engineering
Performance prediction
A priori and a posteriori
Bayes error rate
020201 artificial intelligence & image processing
Artificial intelligence
business
Classifier (UML)
computer
Software
Mathematics
Subjects
Details
- ISSN :
- 15730565 and 08856125
- Volume :
- 107
- Database :
- OpenAIRE
- Journal :
- Machine Learning
- Accession number :
- edsair.doi...........fb4b1cb9d6a6fc9abc9631d3df4bd0e3