ABSTRACT
In recent times, feature extraction attracted much attention in machine learning and pattern recognition fields. This paper extends and improves a scheme for linear feature extraction that can be used in supervised multi-class classification problems. Inspired by recent frameworks for robust sparse LDA and Inter-class sparsity, we propose a unifying criterion able to retain the advantages of these two powerful linear discriminant methods. We introduce an iterative alternating minimization scheme in order to estimate the linear transformation and the orthogonal matrix. The linear transformation is efficiently updated via the steepest descent gradient technique. The proposed framework is generic in the sense that it allows the combination and tuning of other linear discriminant embedding methods. We used our proposed method to fine tune the linear solutions delivered by two recent linear methods: RSLDA and RDA_FSIS. Experiments have been conducted on public image datasets of different types including objects, faces, and digits. The proposed framework compared favorably with several competing methods.
Subject(s)
Algorithms , Pattern Recognition, Automated/trends , Supervised Machine Learning/trends , Discriminant Analysis , Machine Learning/trends , Pattern Recognition, Automated/methodsABSTRACT
Linear Discriminant Analysis (LDA) and its variants are widely used as feature extraction methods. They have been used for different classification tasks. However, these methods have some limitations that need to be overcome. The main limitation is that the projection obtained by LDA does not provide a good interpretability for the features. In this paper, we propose a novel supervised method used for multi-class classification that simultaneously performs feature selection and extraction. The targeted projection transformation focuses on the most discriminant original features, and at the same time, makes sure that the transformed features (extracted features) belonging to each class have common sparsity. Our proposed method is called Robust Discriminant Analysis with Feature Selection and Inter-class Sparsity (RDA_FSIS). The corresponding model integrates two types of sparsity. The first type is obtained by imposing the â2,1 constraint on the projection matrix in order to perform feature selection. The second type of sparsity is obtained by imposing the inter-class sparsity constraint used for ensuring a common sparsity structure in each class. An orthogonal matrix is also introduced in our model in order to guarantee that the extracted features can retain the main variance of the original data and thus improve the robustness to noise. The proposed method retrieves the LDA transformation by taking into account the two types of sparsity. Various experiments are conducted on several image datasets including faces, objects and digits. The projected features are used for multi-class classification. Obtained results show that the proposed method outperforms other competing methods by learning a more compact and discriminative transformation.
Subject(s)
Algorithms , Discriminant Analysis , Pattern Recognition, Automated/methods , Databases, Factual/trends , Humans , Pattern Recognition, Automated/trendsABSTRACT
Natural killer (NK) cells are expanded in chronic myeloid leukemia (CML) patients on tyrosine kinase inhibitors (TKI) and exert cytotoxicity. The inherited repertoire of killer immunoglobulin-like receptors (KIR) may influence response to TKI. We investigated the impact of KIR-genotype on outcome in 166 chronic phase CML patients on first-line imatinib treatment. We validated our findings in an independent patient group. On multivariate analysis, KIR2DS1 genotype (RR=1.51, P=0.03) and Sokal risk score (low-risk RR=1, intermediate-risk RR=1.53, P=0.04, high-risk RR=1.69, P=0.034) were the only independent predictors for failure to achieve complete cytogenetic response (CCyR). Furthermore, KIR2DS1 was the only factor predicting shorter progression-free (PFS) (RR=3.1, P=0.03) and overall survival (OS) (RR=2.6, P=0.04). The association between KIR2DS1 and CCyR, PFS and OS was validated by KIR genotyping in 174 CML patients on first-line imatinib in the UK multi-center SPIRIT-1 trial; in this cohort, KIR2DS1(+) patients had significantly lower 2-year probabilities of achieving CCyR (76.9 vs 87.9%, P=0.003), PFS (85.3 vs 98.1%, P=0.007) and OS (94.4 vs 100%, P=0.015) than KIR2DS1(-) patients. The impact of KIR2DS1 on CCyR was greatest when the ligand for the corresponding inhibitory receptor, KIR2DL1, was absent (P=0.00006). Our data suggest a novel role for KIR-HLA immunogenetics in CML patients on TKI.