Classification of psychophysiological conditions using k-Nearest Neighbour

Author

Aris, Siti Armiza Mohd ; Taib, M.N.

Author_Institution

UTM Razak Sch. of Eng. & Adv. Technol., Univ. Teknol. Malaysia, Kuala Lumpur, Malaysia

fYear

2013

fDate

19-20 Aug. 2013

Firstpage

103

Lastpage

108

Abstract

In this study, the approximate behaviour during resting state and mental task state are predicted to be grouped via asymmetry score. This is realized by applying an advance signal processing technique via alpha spectral. Subtractive clustering is utilized in order to confirm the maximum number of cluster center existed in the combined features of resting state and mental task state. Result showed five cluster centers presented within the data. Hence, the FCM algorithm is sets to generate five clusters, four clusters and three clusters. kNN is chosen to grade the data set of cluster and result showed that three clusters give 100% classification.

Keywords

electroencephalography; fuzzy set theory; learning (artificial intelligence); medical signal processing; neurophysiology; pattern clustering; signal classification; EEG signal; FCM algorithm; advance signal processing technique; alpha spectral; asymmetry score; electroencephalographic signal; fuzzy c-means algorithm; k-nearest neighbour; mental task state; psychophysiological conditions classification; resting state; subtractive clustering; Clustering algorithms; Conferences; Electrodes; Electroencephalography; Euclidean distance; Indexes; Systems engineering and theory; Asymmetry Score; EEG; FCM; Subtractive Clustering; kNN;

fLanguage

English

Publisher

ieee

Conference_Titel

System Engineering and Technology (ICSET), 2013 IEEE 3rd International Conference on

Conference_Location

Shah Alam

Print_ISBN

978-1-4799-1028-1

Type

conf

DOI

10.1109/ICSEngT.2013.6650152

Filename

6650152