Data classification with binary response through the Boosting algorithm and logistic regression
详细信息 查看全文
- 作者：Fortunato S. de Menezes ; a ; b ; fmenezes@dfi.ufla.br" class="auth_mail" title="E-mail the corresponding author ; fsdmenezes@gmail.com" class="auth_mail" title="E-mail the corresponding author ; Gilberto R. Liskab ; gilbertoliska@unipampa.edu.br" class="auth_mail" title="E-mail the corresponding author ; Marcelo A. Cirillob ; macufla@dex.ufla.br" class="auth_mail" title="E-mail the corresponding author ; Má ; rio J.F. Vivancob ; ferrua@dex.ufla.br" class="auth_mail" title="E-mail the corresponding author
- 关键词：Boosting algorithm ; Data classification ; Logistic regression ; Information criteria ; AIC ; BIC ; Selection of models ; Monte Carlo Simulation
- 刊名：Expert Systems with Applications
- 出版年：2017
- 出版时间：1 March 2017
- 年：2017
- 卷：69
- 期：Complete
- 页码：62-73
- 全文大小：1089 K
文摘
- class="listitem" id="list_celist0002">
- class="label">•
Review of AIC and BIC information criteria focused on binary data classification.
- class="label">•
Usual data classification is presented with its drawbacks (i.e., low performance).
- class="label">•
Boosting algorithm showed enhanced results supported by MC simulation.
- class="label">•
Hosmer–Lemeshow test sets the partition of the training(test) for classification.
- class="label">•
CHD disease classification is performed with Boosting showing its high performance.