Data classification and analysis tool (DCAT)


Ing. Jiří Mekyska, prof. Ing. Zdeněk Smékal, CSc.


It is possible to download software here.

Publication to be cited

ELIÁŠOVÁ, I.; MEKYSKA, J.; KOŠŤÁLOVÁ, M.; MAREČEK, R.; SMÉKAL, Z.; REKTOROVÁ, I. Acoustic evaluation of short-term effects of repetitive transcranial magnetic stimulation on motor aspects of speech in Parkinson’s disease. JOURNAL OF NEURAL TRANSMISSION, 2013, vol. 120, no. 4, pp. 597-605. ISSN: 0300-9564.


Data analysis and classification tool (DCAT) is a set of functions that can be used for classification in MATLAB environment. This toolbox covers the whole concept of data classification including several techniques of data split, data normalization, feature selection (based on filtering and wrapping methods), classification, score fusion and classifiers evaluation. This toolbox is able to process the results generated by NDAT (Neurological Disorder Analysis Tool) but it can process any other data with a format specified by DCAT as well. In the current version DCAT provides 6 different classification techniques (Support Vector Machines, Naive Bayes Networks, Discriminant Analysis, k-Nearest Neighbor, Classification Trees and Gaussian Mixture Models), 6 feature selection methods (Conditional Mutual Info Maximization, Min-Redundancy Max-Relevance, Joint Mutual Information, Double Input Symmetrical Relevance, Conditional Redundancy and Sequential Floating Feature Selection) and several evaluation methods (confusion matrix, overall accuracy, particular accuracies, sensitivity, specificity, equal error rate, minimum of detection cost function and area under ROC curve). The DCAT can export a table with the selected evaluation function values and a table with the statistics about the feature selection.

This software uses DETware toolbox that can be downloaded under link and toolbox FEAST (A Feature Selection Toolbox for C and Matlab) that can be downloaded under link The software will not work properly without these toolboxes. Please before use, download these toolboxes end extract their functions directly to a folder toolbox/DETware and toolbox/FEAST, respectively. Then compile all *.c files in folder toolbox\FEAST\MIToolbox and toolbox\FEAST\FSToolbox using MATLAB command “mex”. To test the CDAT please run file demo.m and check if the test passed without any errors. You can also use this file as a starting template and modify it for your data. To get more information about the particular function just type “help name_of_the_function”. Using this command you can also get info about the input and output variables.


This work was supported by projects VG20102014033, NT13499 and FEKT-S-11-17. The described research was performed in laboratories supported by the SIX project; the registration number CZ.1.05/2.1.00/03.0072, the operational program Research and Development for Innovation.


To negotiate the license terms of use of this software please contact the responsible person Ing. Jiri Kouril at Technology Transfer Office, Brno University of Technology, Kounicova 966/67a, Veveří, 60200, Brno, Czech Republic,