Title: Gene expression data classification with robust sparse logistic regression using fused regularisation
Authors: Kampa Lavanya; Pemula Rambabu; G. Vijay Suresh; Rahul Bhandari
Addresses: Department of Computer Science and Engineering, University College of Sciences, Acharya Nagarjuna University, Nagarjuna Nagar, Guntur, India ' Department of Computer Science and Engineering, RAGHU Engineering College, Visakhapatnam, Andhra Pradesh, India ' Department of Computer Science and Engineering, Lakireddy Bali Reddy College of Engineering, Mylavaram, Krishna District, AP, 521230, India ' Department of Computer Science and Engineering, Chandigarh University, Mohali (Punjab)-140413, India
Abstract: Microarray technology has become popular and is extensively used for gene classification. It is essential to identify a proper set of gene expressions that help to classify cancer data. However, microarray data comprises large number of genes with small set of samples. A penalised logistic regression (PLR) is good for variable selection in high dimensional microarray data. The techniques like Lasso, ridge and elastic net are suitable to reduce irrelevant features. However, they failed to produce properties like oracle property and sparsity resulted over fitting. To retain sparsity and oracle property, the weighted L1 and L2 penalties are used in logistic regression for gene expression data. In this paper, a new fused logistic regression (FLR) has been introduced using weighted L1 and L2 penalties for better gene selection. Regression algorithms were tested over the simulated as well as the real gene data sets.
Keywords: microarray data; regularisation; feature selection; sparse logistic regression; robust Lasso; penalised logistic regression; PLR; fused logistic regression; FLR; coordinate descent algorithm; CDA.
International Journal of Ad Hoc and Ubiquitous Computing, 2023 Vol.42 No.4, pp.281 - 291
Accepted: 03 Aug 2022
Published online: 21 Apr 2023 *