A0984
Title: A portmanteau local feature discrimination approach to the classification with high-dimensional matrix-variate data
Authors: Shan Luo - Shanghai Jiao Tong University (China) [presenting]
Zehua Chen - National University of Singapore (Singapore)
Zengchao Xu - Shanghai Normal University (China)
Abstract: Matrix-variate data arise in many scientific fields, such as face recognition, medical imaging, etc. Matrix data contain important structure information, which can be ruined by vectorization. Methods incorporating the structure information into analysis have significant advantages over vectorization approaches. The focus is on the problem of two-class classification with high-dimensional matrix-variate data and the proposal of a novel portmanteau-local-feature discrimination (PLFD) method. The method first identifies local discrimination features of the matrix variate and then pools them together to construct a discrimination rule. The theoretical properties of the PLFD method are investigated and its asymptotic optimality is established. Extensive numerical studies are carried out, including simulation and real data analysis, to compare this method with other methods available in the literature, which demonstrate that the PLFD method has a great advantage over the other methods in terms of misclassification rate.
