266 search results - page 2 / 54 » Is Combining Classifiers Better than Selecting the Best One |
PR
2008
13 years 6 months ago
2008
In handwritten pattern recognition, the multiple classifier system has been shown to be useful for improving recognition rates. One of the most important tasks in optimizing a mul...
ICML
1998
IEEE
14 years 7 months ago
1998
IEEE
In a recent paper, Friedman, Geiger, and Goldszmidt [8] introduced a classifier based on Bayesian networks, called Tree Augmented Naive Bayes (TAN), that outperforms naive Bayes a...
IDA
2007
Springer
14 years 1 months ago
2007
Springer
Random forests are one of the best performing methods for constructing ensembles. They derive their strength from two aspects: using random subsamples of the training data (as in b...
MLMI
2005
Springer
14 years 13 days ago
2005
Springer
Acoustic events produced in meeting-room-like environments may carry information useful for perceptually aware interfaces. In this paper, we focus on the problem of combining diffe...
WSOM
2009
Springer
13 years 11 months ago
2009
Springer
Abstract. The distribution of a class of objects, such as images depicting a specific topic, can be studied by observing the best-matching units (BMUs) of the objects’ feature v...