Assessing the efficiency of multilayer feed-forward neural network model: application to body mass index data
A variety of statistical approaches have been used to find the directed dependencies among a set of interest variables and to identify the associated important factors. Among the most popular methods are proportional hazard regression and logistic regression. The aim of the current study is to sugge...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Published: |
International Digital Organization for Scientific Information (I D O S I)
2011
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/44749/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | A variety of statistical approaches have been used to find the directed dependencies among a set of interest variables and to identify the associated important factors. Among the most popular methods are proportional hazard regression and logistic regression. The aim of the current study is to suggest another approach by using a multilayer feed-forward neural network model (MLFF). Using body mass index (BMI) as the dependent variable, we identify its related and appropriate independent variables. In this study we put forth two MLFF models. Model 1 is where all the independent variables as identified in the literature are included, while Model 2 is where only variables found significance as a result from a multiple linear regression (MLR) analysis are included in the model. Analyses were done by using SPSS and MATLAB packages. As a result of the study, we found that the best MLFF model was the model which considered the input variables based on selection criteria for regression. |
|---|