Intelligent feature subset selection with unspecified number for body fat prediction based on binary-GA and Fuzzy-Binary-GA

Author

Keivanian, Farshid ; Mehrshad, Nasser

Author_Institution

Dept. of Electr. & Comput. Eng., Univ. of Birjand, Birjand, Iran

fYear

2015

fDate

11-12 March 2015

Firstpage

1

Lastpage

7

Abstract

Knowing the body fat is an extremely important issue since it affects everyone´s health. Although there are several ways to measure the body fat percentage (BFP), the accurate methods are often associated with hassle and/or high costs. Therefore, certain measurements or explanatory variables are used to predict the BFP. This study proposes an intelligent feature subset selection approach with unspecified number of features based on Binary GA and Fuzzy Binary GA algorithms to discover most important variable or feature and facilitate an artificial neural network (ANN) classifier model which is applied for body fat prediction (BFP). The proposed forecasting model is able to effectively predict the BFP with error of ± 3.64031% and the most effective feature of forearm circumference among total twelve features by using Fuzzy Binary GA.

Keywords

fats; feature selection; fuzzy set theory; genetic algorithms; health care; medical computing; neural nets; pattern classification; ANN classifier model; BFP; artificial neural network classifier model; body fat percentage; body fat prediction; forearm circumference; forecasting model; fuzzy binary GA algorithms; genetic algorithm; intelligent feature subset selection approach; Artificial neural networks; Cost function; Forecasting; Mathematical model; Predictive models; Radio frequency; Training; Binary GA; Fuzzy Binary GA; artificial neural network (ANN); body fat prediction; intelligent feature subset selection;

fLanguage

English

Publisher

ieee

Conference_Titel

Pattern Recognition and Image Analysis (IPRIA), 2015 2nd International Conference on

Conference_Location

Rasht

Print_ISBN

978-1-4799-8444-2

Type

conf

DOI

10.1109/PRIA.2015.7161651

Filename

7161651