A Comprehensive Review on Discriminant Analysis for Addressing Challenges of Class-Level Limitations, Small Sample Size, and Robustness

被引：1
|
作者：
Qu, Lingxiao [1 ]
Pei, Yan [2 ]
机构：
[1] Univ Aizu, Grad Sch Comp Sci & Engn, Aizu Wakamatsu 9658580, Japan
[2] Univ Aizu, Comp Sci Div, Aizu Wakamatsu 9658580, Japan
来源：
PROCESSES | 2024年 / 12卷 / 07期
关键词：
discriminant analysis; small sample size; singularity; multi-modality; kernel method; PRINCIPAL COMPONENT ANALYSIS; FACE RECOGNITION; FEATURE-EXTRACTION; DIMENSIONALITY REDUCTION; LP-NORM; FAST IMPLEMENTATION; NULL SPACE; CLASSIFICATION; L1-NORM; MODEL;
D O I：
10.3390/pr12071382
中图分类号：
TQ [化学工业];
学科分类号：
0817 ;
摘要：
The classical linear discriminant analysis (LDA) algorithm has three primary drawbacks, i.e., small sample size problem, sensitivity to noise and outliers, and inability to deal with multi-modal-class data. This paper reviews LDA technology and its variants, covering the taxonomy and characteristics of these technologies and comparing their innovations and developments in addressing these three shortcomings. Additionally, we describe the application areas and emphasize the kernel extensions of these technologies to solve nonlinear problems. Most importantly, this paper presents perspectives on future research directions and potential research areas in this field.
引用
收藏
页数：32
相关论文
共 22 条
  • [21] A novel one-parameter regularized linear discriminant analysis for solving small sample size problem in face recognition
    Chen, WS
    Yuen, PC
    Huang, J
    Dai, DQ
    ADVANCES IN BIOMETRIC PERSON AUTHENTICATION, PROCEEDINGS, 2004, 3338 : 320 - 329
  • [22] Addressing the Sample Preparation Challenges in Failure Analysis of Wafer Level Chip Scale Package Mounted Inside a Customer Camera Module
    Lagar, Jason H.
    Sia, Rudolf A.
    Proceedings of the 2016 IEEE 23rd International Symposium on the Physical and Failure Analysis of Integrated Circuits (IPFA), 2016, : 132 - 135
123