EEG-based biometric authentication modelling using incremental fuzzy-rough nearest neighbour technique
This paper proposes an Incremental Fuzzy-Rough Nearest Neighbour (IncFRNN) technique for biometricauthentication modelling using feature extracted visual evoked. Only small training set is needed for model initialisation. Theembedded heuristic update method adjusts the knowledge gr...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | en |
| Published: |
The Institution of Engineering and Technology
2018
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/39184/1/EEG-based%20biometric%20authenticationmodelling%20using%20incremental.pdf http://ir.unimas.my/id/eprint/39184/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This paper proposes an Incremental Fuzzy-Rough Nearest Neighbour (IncFRNN) technique for biometricauthentication modelling using feature extracted visual evoked. Only small training set is needed for model initialisation. Theembedded heuristic update method adjusts the knowledge granules incrementally to maintain all representativeelectroencephalogram (EEG) signal patterns and eliminate those rarely used. It reshapes the personalized knowledge granulesthrough insertion and deletion of a test object, based on similarity measures. A predefined window size can be used to reducethe overall processing time. This proposed algorithm was verified with test data from 37 healthy subjects. Signal pre-processingsteps on segmentation, filtering and artefact rejection were carried out to improve the data quality before model building. Theexperimental paradigm was designed in three different conditions to evaluate the authentication performance of the IncFRNNtechnique against the benchmarked incremental K-Nearest Neighbour (KNN) technique. The performance was measured interms of accuracy, area under the Receiver Operating Characteristic (ROC) curve (AUC) and Cohen's Kappa coefficient. Theproposed IncFRNN technique is proven to be statistically better than the KNN technique in the controlled window sizeenvironment. Future work will focus on the use of dynamic data features to improve the robustness of the proposed model. |
|---|