A rule based approach to classification of EEG datasets: A comparison between ANFIS and rough sets

Author

Jahankhani, Pari ; Revett, Kenneth ; Kodogiannis, Vassilis

Author_Institution

Sch. of Comput. Sci., Univ. of Westminster, London

fYear

2008

fDate

25-27 Sept. 2008

Firstpage

157

Lastpage

160

Abstract

This paper compares two different rule based classification methods in order to evaluate their relative efficiency with respect to classification accuracy and the caliber of the resulting rules. Specifically, the application of adaptive neuro-fuzzy inference system (ANFIS) and rough sets were deployed on a complete dataset consisting of electroencephalogram (EEG) data. The results indicate that both were able to classify this dataset accurately and the number of rules were similar in both cases, provided the dataset was pre-processed using PCA in the case of ANFIS.

Keywords

electroencephalography; fuzzy neural nets; fuzzy reasoning; knowledge based systems; medical computing; pattern classification; principal component analysis; rough set theory; EEG dataset classification; adaptive neuro-fuzzy inference system; electroencephalogram data; principal component analysis; rough set; rule based classification; Adaptive systems; Computer science; Discrete wavelet transforms; Electroencephalography; Epilepsy; Fuzzy neural networks; Neural networks; Principal component analysis; Rough sets; Wavelet coefficients; Neuro-fuzzy systems; PCA; Rough sets; electroencephalography; wavelets;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Network Applications in Electrical Engineering, 2008. NEUREL 2008. 9th Symposium on

Conference_Location

Belgrade

Print_ISBN

978-1-4244-2903-5

Electronic_ISBN

978-1-4244-2904-2

Type

conf

DOI

10.1109/NEUREL.2008.4685599

Filename

4685599