Mixture factor analysis with distance metric constraint for dimensionality reduction
作者：
Highlights：
• We propose a supervised nonlinear DR for high-dimensional data classification.
• Our method is a combination of MFA and Mahalanobis distance metric learning.
• The log-likelihood function of MFA and DMC loss function are jointly optimized.
• Our method can accurately describe the data and enhance the classification results.
• Experimental results show our method performs better than other related methods.
摘要
•We propose a supervised nonlinear DR for high-dimensional data classification.•Our method is a combination of MFA and Mahalanobis distance metric learning.•The log-likelihood function of MFA and DMC loss function are jointly optimized.•Our method can accurately describe the data and enhance the classification results.•Experimental results show our method performs better than other related methods.
论文关键词：Dimensionality reduction,Mixture factor analysis,Distance metric constraint,Classification
论文评审过程：Received 29 June 2020, Revised 25 June 2021, Accepted 3 July 2021, Available online 30 July 2021, Version of Record 5 August 2021.
论文官网地址：https://doi.org/10.1016/j.patcog.2021.108156
