東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Assessing adverse birth outcomes via...

Kitsanta, Panagiota.

FindBook

Google Book

Amazon

博客來

Assessing adverse birth outcomes via classification trees.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Assessing adverse birth outcomes via classification trees./
作者:	Kitsanta, Panagiota.
面頁冊數:	112 p.
附註:	Source: Dissertation Abstracts International, Volume: 64-07, Section: B, page: 3354.
Contained By:	Dissertation Abstracts International64-07B.
標題:	Statistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3098386

Assessing adverse birth outcomes via classification trees.
Kitsanta, Panagiota.

Assessing adverse birth outcomes via classification trees. - 112 p.

Source: Dissertation Abstracts International, Volume: 64-07, Section: B, page: 3354.

Thesis (Ph.D.)--The Florida State University, 2003.

The purpose of this study is to investigate the geographical distribution of adverse birth outcomes using classification trees and logistic regression models, evaluate these methods, and examine the stability of classification trees. We use ROC curves to evaluate the predictive performance of tree-structured models and logistic regression. The stability of tree classifiers is assessed by an independent-test sample technique and the bootstrapping approach. These issues are addressed by using data extracted from birth certificates registered during the years of 1998 and 1999 in Florida.Subjects--Topical Terms:

517247
Statistics.

Assessing adverse birth outcomes via classification trees.
LDR:02706nmm 2200289 4500 001 1860504
005 20041028080311.5
008 130614s2003 eng d
035 $a (UnM)AAI3098386
035 $a AAI3098386
040 $a UnM $c UnM
100 1 $a Kitsanta, Panagiota. $3 1948137
245 1 0 $a Assessing adverse birth outcomes via classification trees.
300 $a 112 p.
500 $a Source: Dissertation Abstracts International, Volume: 64-07, Section: B, page: 3354.
500 $a Major Professor: Myles Hollander.
502 $a Thesis (Ph.D.)--The Florida State University, 2003.
520 $a The purpose of this study is to investigate the geographical distribution of adverse birth outcomes using classification trees and logistic regression models, evaluate these methods, and examine the stability of classification trees. We use ROC curves to evaluate the predictive performance of tree-structured models and logistic regression. The stability of tree classifiers is assessed by an independent-test sample technique and the bootstrapping approach. These issues are addressed by using data extracted from birth certificates registered during the years of 1998 and 1999 in Florida.
520 $a Findings indicate that non-healthy White, Hispanic or Other non-white women had a significantly higher chance of an adverse birth if they had no previous births or they were not married. Healthy Hispanic and White women were at greater risk of an adverse birth if they gained less than 20lbs during pregnancy, were not married, did not have any prior pregnancies, and either had postgraduate or less than high school education. The independent-tested, rule A (equal misclassification costs and priors) and B (unequal misclassification costs and equal priors) classifiers proved to be stable overall. Rule B classifiers were more stable compared to rule A models. In contrast, when rule A tree models were applied to bootstrap replicates, we observed that these models were unstable. Stability continued to exist among rule B bootstrap classifiers. Based on comparisons between classification trees and logistic regression models, we cannot conclude that one method outperformed the other.
520 $a The information generated from this study can be utilized by officials who are interested in allocating resources, and designing and implementing programs that address the needs of each region more effectively within the state of Florida.
590 $a School code: 0071.
650 4 $a Statistics. $3 517247
650 4 $a Biology, Biostatistics. $3 1018416
690 $a 0463
690 $a 0308
710 2 0 $a The Florida State University. $3 1017727
773 0 $t Dissertation Abstracts International $g 64-07B.
790 1 0 $a Hollander, Myles, $e advisor
790 $a 0071
791 $a Ph.D.
792 $a 2003
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3098386