Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Editorial

Editorial Journal of Classification 32: 1-2 (2015) DOI: 10.1007/s00357- 015- 9 176 - 0 The first issue of this year starts with a relatively new type of supervised classification that has become especially relevant in the upcoming field of connectomics, the study of networks in nervous systems at the level of synaptic connections. As explained by Vogelstein and Priebe in the lead article, the data in graph classification are a collection of networks or graphs rather than a collection of points or vectors, as is usually the case. The authors develop a random graph model in which theoretical and computational questions can be answered, for example about the consistency of graph classifiers, and – as another example – whether for networks with a moderately large number of vertices (say, more than thirty) there exist classifiers with the property that they do not require more space than there are atoms in the universe. Next we turn to classification trees, for which Bar-Hen, Gey and Poggi look at the problem of evaluating the sensitivity of the results due to variability in the data. They discuss several influence measures and derive distributional results for them. These can be used to assess the stability http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Classification Springer Journals

Editorial

Heiser, Willem
Journal of Classification , Volume 32 (1) – Apr 18, 2015
Download PDF
2 pages

Loading next page...
 
/lp/springer-journals/editorial-Hdl5ZtAQvN

References (0)

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Springer Journals
Copyright
Copyright © 2015 by Classification Society of North America
Subject
Statistics; Statistical Theory and Methods; Pattern Recognition; Bioinformatics; Signal, Image and Speech Processing; Psychometrics; Marketing
ISSN
0176-4268
eISSN
1432-1343
DOI
10.1007/s00357-015-9176-0
Publisher site
See Article on Publisher Site

Abstract

Journal of Classification 32: 1-2 (2015) DOI: 10.1007/s00357- 015- 9 176 - 0 The first issue of this year starts with a relatively new type of supervised classification that has become especially relevant in the upcoming field of connectomics, the study of networks in nervous systems at the level of synaptic connections. As explained by Vogelstein and Priebe in the lead article, the data in graph classification are a collection of networks or graphs rather than a collection of points or vectors, as is usually the case. The authors develop a random graph model in which theoretical and computational questions can be answered, for example about the consistency of graph classifiers, and – as another example – whether for networks with a moderately large number of vertices (say, more than thirty) there exist classifiers with the property that they do not require more space than there are atoms in the universe. Next we turn to classification trees, for which Bar-Hen, Gey and Poggi look at the problem of evaluating the sensitivity of the results due to variability in the data. They discuss several influence measures and derive distributional results for them. These can be used to assess the stability

Journal

Journal of Classification – Springer Journals

Published: Apr 18, 2015

There are no references for this article.