東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Bootstrap techniques for statistical...

Wang, Qun.

Linked to FindBook

Google Book

Amazon

博客來

Bootstrap techniques for statistical pattern recognition.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Bootstrap techniques for statistical pattern recognition./
Author:	Wang, Qun.
Description:	138 p.
Notes:	Source: Masters Abstracts International, Volume: 39-01, page: 0245.
Contained By:	Masters Abstracts International39-01.
Subject:	Computer Science. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=MQ52407
ISBN:	0612524078

Bootstrap techniques for statistical pattern recognition.
Wang, Qun.

Bootstrap techniques for statistical pattern recognition. - 138 p.

Source: Masters Abstracts International, Volume: 39-01, page: 0245.

Thesis (M.Sc.)--Carleton University (Canada), 2000.

This thesis studies the application of the bootstrap techniques to three fields of statistical pattern recognition: the estimation of the Bhattacharyya bound, the estimation of the error rate of a classifier and the classifier design. To initiate the study, the bootstrap technique, and its concepts, algorithms and properties are explained.

ISBN: 0612524078Subjects--Topical Terms:

626642
Computer Science.

Bootstrap techniques for statistical pattern recognition.
LDR:02519nam 2200289 a 45 001 929266
005 20110427
008 110427s2000 eng d
020 $a 0612524078
035 $a (UnM)AAIMQ52407
035 $a AAIMQ52407
040 $a UnM $c UnM
100 1 $a Wang, Qun. $3 1250509
245 1 0 $a Bootstrap techniques for statistical pattern recognition.
300 $a 138 p.
500 $a Source: Masters Abstracts International, Volume: 39-01, page: 0245.
502 $a Thesis (M.Sc.)--Carleton University (Canada), 2000.
520 $a This thesis studies the application of the bootstrap techniques to three fields of statistical pattern recognition: the estimation of the Bhattacharyya bound, the estimation of the error rate of a classifier and the classifier design. To initiate the study, the bootstrap technique, and its concepts, algorithms and properties are explained.
520 $a In the application of the bootstrap to estimating the Bhattacharyya bound, it is first argued, both theoretically and experimentally that the estimate of the Bhattacharyya bound is usually biased. Two types of the bootstrap estimators of the Bhattacharyya bound, the direct and indirect bias correction estimators, are suggested. It is proved experimentally that the suggested bootstrap estimators of the Bhattacharyya bound are superior to the general estimator.
520 $a In the application of the bootstrap to the error rate estimation, the previous works are reviewed, and the related algorithms are illustrated with the simulation results. These algorithms are catalogued against the real-sample algorithm, as they use only the original training samples to estimate the error rate. The pseudo-sample algorithms of the error rate estimation are then proposed. These use the bootstrap technique to generate pseudosamples for estimating the error rate. Comparisons based on the experiments prove that the pseudo-sample algorithms perform better than most of the other algorithms.
520 $a In the application of the bootstrap to classifier design, the mixed-sample algorithm is introduced. It overcomes the limitation of the training sample size by adding pseudo-samples to the original training samples for classifier design. The experimental results prove that this strategy is successful.
590 $a School code: 0040.
650 4 $a Computer Science. $3 626642
650 4 $a Statistics. $3 517247
690 $a 0463
690 $a 0984
710 2 0 $a Carleton University (Canada). $3 1018407
773 0 $t Masters Abstracts International $g 39-01.
790 $a 0040
791 $a M.Sc.
792 $a 2000
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=MQ52407