Title: A comparative study on machine classification model in lung cancer cases analysis

Authors: Jing Li; Zhisheng Zhao; Yang Liu; Jie Li; Zhiwei Cheng; Xiaozheng Wang

Addresses: School of Information Science and Engineering, Hebei North University, Zhangjiakou, China ' School of Information Science and Engineering, Hebei North University, Zhangjiakou, China ' School of Information Science and Engineering, Hebei North University, Zhangjiakou, China ' China-Japan Friendship Hospital, No. 2 East Yinghuayuan Street, Chaoyang District, Beijing, China ' School of Information Science and Engineering, Hebei North University, Zhangjiakou, China ' Zhangjiakou University, No. 19 Pingmen Road, Qiaoxi District, Zhangjiakou, China

Abstract: Due to the differences of machine classification models in the application of medical data, this paper selected different classification methods to study lung cancer data collected from HIS system with plenty of experiment and analysis, applying the R language on decision tree algorithm, bagging algorithm, Adaboost algorithm, conditions decision tree, random forests, naive Bayes, and neural network algorithm for lung cancer data analysis, in order to explore the advantages and disadvantages of each machine classification algorithm. The results confirmed that in lung cancer data research, naive Bayes, Adaboost algorithm and neural network algorithm have relatively high accuracy, with a good diagnostic performance.

Keywords: machine classification model; lung cancer; cross validation; R language; Adaboost algorithm; neural network; decision tree; conditions inference tree; random forest; naive Bayes; bagging algorithm.

DOI: 10.1504/IJASS.2017.088906

International Journal of Applied Systemic Studies, 2017 Vol.7 No.1/2/3, pp.13 - 29

Received: 05 Oct 2016
Accepted: 05 Apr 2017
Published online: 02 Jan 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article