文献类型：期刊文献

英文题名：Adaptive multinomial regression with overlapping groups for multi-class classification of lung cancer

作者：Li, Juntao[1];Wang, Yanyan[1];Song, Xuekun[2];Xiao, Huimin[3]

第一作者：Li, Juntao

通讯作者：Li, JT[1]

机构：[1]Henan Normal Univ, Coll Math & Informat Sci, Xinxiang 453007, Peoples R China;[2]Henan Univ Chinese Med, Coll Informat Technol, Zhengzhou 450046, Henan, Peoples R China;[3]Henan Univ Econ & Law, Coll Comp & Informat Engn, Zhengzhou 450002, Henan, Peoples R China

第一机构：Henan Normal Univ, Coll Math & Informat Sci, Xinxiang 453007, Peoples R China

通讯机构：[1]corresponding author), Henan Normal Univ, Coll Math & Informat Sci, Xinxiang 453007, Peoples R China.

年份：2018

卷号：100

起止页码：1-9

外文期刊名：COMPUTERS IN BIOLOGY AND MEDICINE

收录：;EI(收录号：20182605375407);Scopus(收录号：2-s2.0-85048994283);WOS:【SCI-EXPANDED(收录号:WOS:000442704300001)】；

基金：This work was supported by the Natural Science Foundation of China (61203293, 61702161, 61602153, 61702164), Scientific and Technological Project of Henan Province (172102210047, 162102310461, 172102310535, 182102210020), Natural Science Foundation of Henan Province (162300410184), Foundation of Henan Educational Committee (18A520015,18A520003, 188510004), Scientific Research Project of Zhengzhou (153PKJGG128), Foundation for University Young Key Teacher of Henan Province (2016GGJS-079).

语种：英文

外文关键词：Multi-class classification; Imbalanced data; Overlapping group lasso; Weighted gene co-expression networks

摘要：Multi-class classification has attracted much attention in cancer diagnosis and treatment and many machine learning methods have emerged for addressing this issue recently. However, class imbalance and gene selection problems occur in classifying lung cancer data. In this paper, an adaptive multinomial regression with a sparse overlapping group lasso penalty is proposed to perform classification and grouped gene selection for lung cancer gene expression data. An overlapped grouping strategy with biological interpretability is proposed, which highlights the importance of gene groups from the minority classes. By using the conditional mutual information, the gene significance within each group is evaluated and the data-driven weights are constructed. Based on the grouping strategy and constructed weights, a regularized adaptive multinomial regression is presented and the solving algorithm is developed, which can not only select the important gene groups for each class in performing multi-class classification, but also adaptively select important genes within each group. The experiment results show that the proposed method significantly outperforms the other 6 methods on classification accuracy, and the selected genes are disease-causing genes for lung cancer.