Biomarker Discovery based on BBHA and AdaboostM1 on Microarray Data for Cancer Classification

Pashaei E., Ozen M., AYDIN N.

38th Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society (EMBC), Florida, United States Of America, 16 - 20 August 2016, pp.3080-3083, (Full Text) identifier identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/embc.2016.7591380
  • City: Florida
  • Country: United States Of America
  • Page Numbers: pp.3080-3083
  • Keywords: Gene selection, AdaboostM1, binary black hole algorithm, cancer classification
  • Kocaeli University Affiliated: No

Abstract

In this paper, a new approach based on Binary Black Hole Algorithm (BBHA) and Adaptive Boosting version M1 (AdaboostM1) is proposed for finding genes that can classify the group of cancers correctly. In this approach, BBHA is used to perform gene selection and AdaboostM1 with 10-fold cross validation is adopted as the classifier. Also, to find the relation between the biomarkers for biological point of view, decision tree algorithm (C4.5) is utilized. The proposed approach is tested on three benchmark microarrays. The experimental results show that our proposed method can select the most informative gene subsets by reducing the dimension of the data set and improve classification accuracy as compared to several recent studies.