Dimensionality Reduction of Multimodal Labeled Data by Local Fisher Discriminant Analysis

By Masashi Sugiyama. Journal of machine learning research, 2007

1,137 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/9214714fad458c45faa353c237400f0aa412dbbc

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Motor Bearing Fault Diagnosis Using Trace Ratio Linear Discriminant Analysis (2014, methodology): “Typical methods include MFA [36], locality sensitive discriminant analysis [39], local Fisher discriminant analysis [40], [41] and robust linearly optimized discriminant analysis [42].”
Change-Point Detection in Time-Series Data by Direct Density-Ratio Estimation (2009, methodology): “A possible measure for this would be to incorporate some dimensionality reduction scheme [9, 4, 29, 32] into the density-ratio estimation framework.”
Projection Metric Learning on Grassmann Manifold with Application to Video based Face Recognition (2015, methodology): “On YTF dataset, since DCC [27], GDA [12] and GGDA [16] are not specifically designed for pair-wise based face verification, we modify their original LDA-like part as its pairwise version (like the two works [33, 25]) by constructing the within-class scatter matrix from intra-class pairs and the betw”
Spectral–Spatial Feature Extraction for Hyperspectral Image Classification: A Dimension Reduction and Deep Learning Approach (2016, methodology): “For evaluating the performance of the proposed SSFC, the classification results of SSFC are compared to other commonly used DR methods and spectral–spatial feature extraction methods, including PCA, LDA, LDE, RLDE, LFDA, NWFE, EMPs [32], and spectral+EMP (SEMP) [33].”
Text Classification Algorithms: A Survey (2019, methodology): “Linear Discriminant Analysis (LDA) linear discriminant analysis (LDA) is a commonly used technique for data classification and dimensionality reduction [76].”
A Two-Phase Test Sample Sparse Representation Method for Use With Face Recognition (2011, methodology): “For example, Sugiyama integrated the ideas of the LDA and locality preserving projection to generate a local transform method for a problem where the samples in a class are multimodal [13].”
Hyperspectral Image Classification—Traditional to Deep Models: A Survey for Future Prospects (2021, background): “For instance, linear discriminant analysis (LDA), local Fisher discriminant analysis (LFDA) [101], local discriminant embedding (LDE) [102] and nonparametric weighted feature extraction (NWFE) [103].”
Locality-Preserving Dimensionality Reduction and Classification for Hyperspectral Image Analysis (2012, background): “LFDA [10] has been recently proposed as an extension to LDA, which, by not restricting the class distributions to be unimodal Gaussian, is expected to outperform LDA significantly for many practical classification situations.”
Graph embedding discriminant analysis on Grassmannian manifolds for improved image set matching (2011, methodology): “To alleviate the above problem, a graph-embedding framework can be used [7, 24, 30].”
Feature Extraction for Hyperspectral Imagery: The Evolution From Shallow to Deep: Overview and Toolbox (2020, methodology): “Essentially, LFDA can be regarded as a weighted LDA by locally weighing Sw and Sb matrices.”
Change-Point Detection in Time-Series Data by Relative Density-Ratio Estimation (2012, methodology): “A possible measure for this would be to incorporate some dimensionality reduction scheme [9, 4, 29, 32] into the density-ratio estimation framework.”
Geometric Mean for Subspace Selection (2009, methodology): “From Tables 2, 3, 4, and 5, it can be concluded that MGMD usually outperforms FLDA, HDA, aPAC, WFLDA, FS-FLDA, LFLDA, HFLDA, and ODA, consistently.”