Subclass-Graph-Embedding

This repository contains Matlab code that performs dimensionality reduction and classification using the Subclass Graph Embedding methodology presented in journal paper "Maronidis, A., Tefas, A., & Pitas, I. (2015). Subclass graph embedding and a marginal fisher analysis paradigm. Pattern Recognition, 48(12), 4024-4035." It also contains code that performs multi-scale Spectral Clustering based on paper "Azran, A., & Ghahramani, Z. (2006, June). Spectral methods for automatic multiscale data clustering. In 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06) (Vol. 1, pp. 190-197). IEEE.".

"osu-svm" toolbox that is publicly available under "BSD license" in the following site https://sourceforge.net/projects/svm/ is mandatory in the case that the end user needs to run SVM classifier.

Contained files:

pca.m
RandomDataGenerator.m
SGE_AdjustParameterCV.m
SGE_Assessment.m
SGE_ClassCentroids.m
SGE_Classification.m
SGE_ConfusionMatrix.m
SGE_CrossDivision.m
SGE_CrossValidation.m
SGE_DimReduction.m
SGE_DivisionIds.m
SGE_Evaluation.m
SGE_FindLocalMax.m
SGE_FindParameters.m
SGE_FindPlausible.m
SGE_GramMatrix.m
SGE_GraphConstruct.m
SGE_LaplacianEigenanalysis.m
SGE_Mapping.m
SGE_Maximin.m
SGE_MultiSpecCluster.m
SGE_NearestCentroid.m
SGE_NearestClusterCentroid.m
SGE_PatternRecognition.m
SGE_Projection.m
SGE_SubclassExtract.m
SGE_SubclassLabels.m