Network-based Dimensionality Reduction and Analysis (NDA)

Network-based dimensionality reduction and analysis in MATLAB

nda_matlab

Network-based dimensionality reduction and analysis in MATLAB

This package provides Network-based dimensionality reduction and analysis.

• Network-based dimensionality reduction and analysis.

• Dimensional reduction.

• Plot and biplot functions

• Data generation

Author

• Zsolt T. Kosztyan

Contributor

• Zsolt T. Kosztyan

Maintainer

• Zsolt T. Kosztyan

Outputs:

Scores: n by m matrix of factor scores, where n is the number of rows in a datasource, m is tne number of latent factors CMTX: m by m factor correlation matrix COMMUNALITY: n by 1 row vector of communalities LOADINGS: s by m matrix of factor loadings, where s is the number of selected indicators LTABLE: s by m table of factor loadings, where s is the number of % selected indicators MEMBERSHIPS: m by 1 vector of membership

Input:

data: n by M matrix/table/structure of data source (mandatory)

Optional input parameters:

XHeader: M by 1 cell array of variable names CorrMethod|cor_method: Correlation method (optional) Pearson|pearson|'1'|1: Pearson's correlation (default) Spearman|spearman|'2'|2: Spearman's correlation Kendall|kendall|'3'|3: Kendall's correlation Distance|distance|'4'|4: Distance correlation -otherwise: 1 (Pearson's correlation) MinCor2|min_R: Minimal square correlation between indicators (default: 0) MinimalCommunity|min_comm: Minimal number of indicators in a community (default: 2) Gamma: Gamma parameter in multiresolution null_modell (default: 1) NullModelType|null_model_type (default: 1); NewmannGrivan|'1'|1: Newmann-Grivan's null modell AvgDet: Null model is the mean of square correlations between indicators MinDet,min_det: Null modell is the specified minimal square correlation (min_det) MinEigCentValue|min_evalue: Minimal EVC value (default: 0.00) MinCommunality|min_communality: Minimal communality value of indicators (default: 0.25) ComCommunalities|com_communalities=0.0: Minimal common communalities RotateMethod: Rotation method (default: none); Biplots: Draw biplots (default: false) cuts: Draw correlation graph with cuts value (default: 0 => No correlation graph)

Usages:

[Scores,CMTX,COMMUNALITY,LOADINGS,LTABLE,MEMBERSHIPS]=nda(data) [Scores,CMTX,COMMUNALITY,LOADINGS,LTABLE,MEMBERSHIPS]=nda(data,Xheader) [Scores,CMTX,COMMUNALITY,LOADINGS,LTABLE,MEMBERSHIPS]=nda(data,Xheader,...)

Examples:

load CWTS_2020 [Scores,CMTX,COMMUNALITY,LOADINGS,LTABLE,MEMBERSHIPS]=nda(CWTS_2020) [Scores,CMTX,COMMUNALITY,LOADINGS,LTABLE,MEMBERSHIPS]=nda(CWTS_2020,'RotationMethod','varimax','MinimalCommunity',3)

Requirements:

Eigenvector centralities (if Matlab release is older than R2020a) (Contributors): Xi-Nian Zuo, Chinese Academy of Sciences, 2010 Rick Betzel, Indiana University, 2012 Mika Rubinov, University of Cambridge, 2015

Modified GenLouvain toolbox (Contributurs): Lucas G. S. Jeub, Marya Bazzi, Inderjit S. Jutla, and Peter J. Mucha, "A generalized Louvain method for community detection implemented in MATLAB," https://github.com/GenLouvain/GenLouvain (2011-2019).