Spatial Weighted Fuzzy Clustering
folder structure:
data contains 3 data sets: ds4.csv is a sythetic 2D dataset for evaluation 2d.csv is a 2D geometallurgical dataset synthetic_bm_small.csv is a full 3D sythetic geometallurgical dataset
#fortran modules utils.f90 utilities, for example, wrapper to random number generator distances.f90 different distances functions to support dimension distances for continuous, categorical and targeted attributes
clustering.f90 fuzzy clustering functions supporting WFC
##How to build fortran extension make
plotting.py utilities for plotting using matplotlib
categorical_utils.py utilities for supporting categorical attributes cluster_utils.py utilities for clustering
graph_labeling.py wrapper to pygco pytho extension which must be separatedly installed
clustering_ga.py GA optimisation for SWFC method (uses deap Python package)
METHOD kmean, pca, wfc DATASET ds4, 2d, bm
boxplot_.py Generates boxplot charts case_study_.py Loads the case study data clustering__.py Performs clustering usign a METHOD to a DATASET plot_.py Plot many charts, see doc in each script spatial_correction_.py Performs Spatial Correction (SWFC method)