Library Header Image
LSE Research Online LSE Library Services

Using a similarity measure for credible classification

Anthony, Martin and Hammer, P. L. and Subasi, E. and Subasi, M. (2005) Using a similarity measure for credible classification. CDAM research report series, CDAM-LSE-2005-22. Centre for Discrete and Applicable Mathematics, London School of Economics and Political Science, London, UK.

Full text not available from this repository.
Identification Number: CDAM-LSE-2005-22


This paper concerns classification by Boolean functions. We investigate the classification accuracy obtained by standard classification techniques on unseen points (elements of the domain, {0, 1}n, for some n) that are similar, in particular senses, to the points that have been observed as training obser- vations. Explicitly, we use a new measure of how similar a point x ∈ {0, 1}n is to a set of such points to restrict the domain of points on which we offer a classification. For points sufficiently dissimilar, no classification is given. We report on experimental results which indicate that the classification ac- curacies obtained on the resulting restricted domains are better than those obtained without restriction. These experiments involve a number of standard data-sets and classification techniques. We also compare the classification ac- curacies with those obtained by restricting the domain on which classification is given by using the Hamming distance.

Item Type: Monograph (Report)
Official URL:
Additional Information: © 2005 the authors
Subjects: Q Science > QA Mathematics
Sets: Departments > Mathematics
Date Deposited: 23 Oct 2008 09:43
Last Modified: 08 Nov 2012 16:09

Actions (login required)

View Item View Item