Validation of Nearest Neighbor Classifiers
We develop a probabilistic bound on the error rate of the nearest neighbor classifier formed from a set of labelled examples. The bound is computed using only the examples in the set. A subset of the examples is used as a validation set to bound the error rate of the classifier formed from the remai...
Main Author: | |
---|---|
Other Authors: | |
Format: | Text |
Language: | English |
Published: |
1998
|
Subjects: | |
Online Access: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.44.1800 http://www.math.urich.edu/~ebax/papers/nearest_neighbor.ps |
id |
ftciteseerx:oai:CiteSeerX.psu:10.1.1.44.1800 |
---|---|
record_format |
openpolar |
spelling |
ftciteseerx:oai:CiteSeerX.psu:10.1.1.44.1800 2023-05-15T17:32:54+02:00 Validation of Nearest Neighbor Classifiers Eric Bax The Pennsylvania State University CiteSeerX Archives 1998 application/postscript http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.44.1800 http://www.math.urich.edu/~ebax/papers/nearest_neighbor.ps en eng http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.44.1800 http://www.math.urich.edu/~ebax/papers/nearest_neighbor.ps Metadata may be used without restrictions as long as the oai identifier remains attached to it. http://www.math.urich.edu/~ebax/papers/nearest_neighbor.ps text 1998 ftciteseerx 2016-01-08T05:11:29Z We develop a probabilistic bound on the error rate of the nearest neighbor classifier formed from a set of labelled examples. The bound is computed using only the examples in the set. A subset of the examples is used as a validation set to bound the error rate of the classifier formed from the remaining examples. Then a bound is computed for the difference in error rates between the original classifier and the reduced classifier. This bound is computed by partitioning the validation set and using each subset to compute bounds for the error rate difference due to the other subsets. 1 Framework Consider the following machine learning framework. There is an unknown boolean-valued target function, and there is a distribution over the input space of the function. For example, the input distribution could consist of typical satellite images of the North Atlantic Ocean, and the target function could be 1 if the image contains a large iceberg and 0 otherwise. We have a set of in-sample data e. Text North Atlantic Unknown |
institution |
Open Polar |
collection |
Unknown |
op_collection_id |
ftciteseerx |
language |
English |
description |
We develop a probabilistic bound on the error rate of the nearest neighbor classifier formed from a set of labelled examples. The bound is computed using only the examples in the set. A subset of the examples is used as a validation set to bound the error rate of the classifier formed from the remaining examples. Then a bound is computed for the difference in error rates between the original classifier and the reduced classifier. This bound is computed by partitioning the validation set and using each subset to compute bounds for the error rate difference due to the other subsets. 1 Framework Consider the following machine learning framework. There is an unknown boolean-valued target function, and there is a distribution over the input space of the function. For example, the input distribution could consist of typical satellite images of the North Atlantic Ocean, and the target function could be 1 if the image contains a large iceberg and 0 otherwise. We have a set of in-sample data e. |
author2 |
The Pennsylvania State University CiteSeerX Archives |
format |
Text |
author |
Eric Bax |
spellingShingle |
Eric Bax Validation of Nearest Neighbor Classifiers |
author_facet |
Eric Bax |
author_sort |
Eric Bax |
title |
Validation of Nearest Neighbor Classifiers |
title_short |
Validation of Nearest Neighbor Classifiers |
title_full |
Validation of Nearest Neighbor Classifiers |
title_fullStr |
Validation of Nearest Neighbor Classifiers |
title_full_unstemmed |
Validation of Nearest Neighbor Classifiers |
title_sort |
validation of nearest neighbor classifiers |
publishDate |
1998 |
url |
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.44.1800 http://www.math.urich.edu/~ebax/papers/nearest_neighbor.ps |
genre |
North Atlantic |
genre_facet |
North Atlantic |
op_source |
http://www.math.urich.edu/~ebax/papers/nearest_neighbor.ps |
op_relation |
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.44.1800 http://www.math.urich.edu/~ebax/papers/nearest_neighbor.ps |
op_rights |
Metadata may be used without restrictions as long as the oai identifier remains attached to it. |
_version_ |
1766131223852220416 |