A `no panacea theorem' for classifier combination

Hu, R. and Damper, R. I. (2008) A `no panacea theorem' for classifier combination Pattern Recognition, 41, (8), pp. 2665-2673. (doi:10.1016/j.patcog.2008.01.022).


Full text not available from this repository.


We introduce the ‘No Panacea Theorem’ (NPT) for multiple classifier combination, previously proved only in the case of two classifiers and two classes. In this paper, we extend the NPT to cases of multiple classifiers and multiple classes. We prove that if the combination function is continuous and diverse, there exists a situation in which the combination algorithm will give very bad performance. The proof relies on constructing ‘pathological’ probability density distributions that have high densities in particular areas such that the combination functions give incorrect classification. Thus, there is no optimal combination algorithm that is suitable in all situations. It can be seen from this theorem that the probability density functions (pdfs) play an important role in the performance of combination algorithms, so studying the pdfs becomes the first step of finding a good combination algorithm. Although devised for classifier combination, the NPT is also relevant to all supervised classification problems

Item Type: Article
Digital Object Identifier (DOI): doi:10.1016/j.patcog.2008.01.022
ISSNs: 0031-3203 (print)
Organisations: Southampton Wireless Group
ePrint ID: 265147
Date :
Date Event
August 2008Published
Date Deposited: 05 Feb 2008 16:15
Last Modified: 17 Apr 2017 19:25
Further Information:Google Scholar
URI: http://eprints.soton.ac.uk/id/eprint/265147

Actions (login required)

View Item View Item