Using a similarity measure for credible classification.

Using a similarity measure for credible classification.
复制标题

DOI:
10.1016/j.dam.2008.04.007
复制
发表时间:
2009-03-06
影响因子:
1.1
通讯作者:
Hammer, P. L.
Hammer, P. L.
中科院分区:
数学3区
文献类型:
--
作者:
Subasi, M.;Subasi, E.;Anthony, M.;Hammer, P. L.

文献摘要

参考文献

相似文献

本文讨论布尔函数分类问题。我们调查的分类精度所获得的标准分类技术上看不见的点(域的元素,{0,1}n,对于一些n)是相似的,在特定的意义上,已经观察到的点作为训练观察。解释,我们使用一个新的措施,如何相似的一个点x ∈ {0,1}n是一组这样的点,以限制域上的点,我们提供了一个分类。对于足够不相似的点,不给出分类。我们报告的实验结果表明,所得到的限制域上的分类精度优于那些没有限制。这些实验涉及一些标准数据集和分类技术。我们还比较了分类精度与通过限制域上的分类是通过使用汉明距离。
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 observations. 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 accuracies 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 accuracies with those obtained by restricting the domain on which classification is given by using the Hamming distance.
DOI: 10.1162/089976601300014493
发表时间: 2001-03-01
期刊: NEURAL COMPUTATION
影响因子: 2.9
作者:
Keerthi, SS;Shevade, SK;Murthy, KRK
通讯作者: Murthy, KRK
DOI: 10.1007/bf02614316
发表时间: 1997-10-01
影响因子: 2.7
作者:
Boros, E;Hammer, PL;Kogan, A
通讯作者: Kogan, A
DOI: 10.1109/69.842268
发表时间: 2000-03-01
影响因子: 8.9
作者:
Boros, E;Hammer, PL;Muchnik, I
通讯作者: Muchnik, I
DOI: 10.1145/76359.76371
发表时间: 1989-10-01
期刊: JOURNAL OF THE ACM
影响因子: 2.5
作者:
BLUMER, A;EHRENFEUCHT, A;WARMUTH, MK
通讯作者: WARMUTH, MK
DOI: 10.1007/bf00153759
发表时间: 1991-01-01
期刊: MACHINE LEARNING
影响因子: 7.5
作者:
AHA, DW;KIBLER, D;ALBERT, MK
通讯作者: ALBERT, MK