Show simple item record

dc.contributor.authorOnaran, İbrahimen_US
dc.contributor.authorİnce, N. Fıraten_US
dc.contributor.authorAbosch, A.en_US
dc.contributor.authorÇetin, A. Enisen_US
dc.coverage.spatialSantander, Spainen_US
dc.date.accessioned2016-02-08T12:10:29Z
dc.date.available2016-02-08T12:10:29Z
dc.date.issued2012en_US
dc.identifier.urihttp://hdl.handle.net/11693/28078
dc.descriptionDate of Conference: 23-26 Sept. 2012en_US
dc.description.abstractCommon Spatial Pattern algorithm (CSP) is widely used in Brain Machine Interface (BMI) technology to extract features from dense electrode recordings by using their weighted linear combination. However, the CSP algorithm, is sensitive to variations in channel placement and can easily overfit to the data when the number of training trials is insufficient. Construction of sparse spatial projections where a small subset of channels is used in feature extraction, can increase the stability and generalization capability of the CSP method. The existing 0 norm based sub-optimal greedy channel reduction methods are either too complex such as Backward Elimination (BE) which provided best classification accuracies or have lower accuracy rates such as Recursive Weight Elimination (RWE) and Forward Selection (FS) with reduced complexity. In this paper, we apply the Oscillating Search (OS) method which fuses all these greedy search techniques to sparsify the CSP filters. We applied this new technique on EEG dataset IVa of BCI competition III. Our results indicate that the OS method provides the lowest classification error rates with low cardinality levels where the complexity of the OS is around 20 times lower than the BE. © 2012 IEEE.en_US
dc.language.isoEnglishen_US
dc.source.title2012 IEEE International Workshop on Machine Learning for Signal Processingen_US
dc.relation.isversionofhttp://dx.doi.org/10.1109/MLSP.2012.6349752en_US
dc.subjectBrain Machine Interfaceen_US
dc.subjectElectroencephalogram (EEG)en_US
dc.subjectOscillating Searchen_US
dc.subjectSparse Filteren_US
dc.subjectAccuracy rateen_US
dc.subjectBackward eliminationen_US
dc.subjectBrain machine interfaceen_US
dc.subjectCardinalitiesen_US
dc.subjectChannel reductionen_US
dc.subjectClassification accuracyen_US
dc.subjectClassification error rateen_US
dc.subjectCommon spatial patternsen_US
dc.subjectData setsen_US
dc.subjectForward selectionen_US
dc.subjectGeneralization capabilityen_US
dc.subjectGreedy searchen_US
dc.subjectIn-channelsen_US
dc.subjectLinear combinationsen_US
dc.subjectOscillating Searchen_US
dc.subjectReduced complexityen_US
dc.subjectSparse Filteren_US
dc.subjectSpatial filtersen_US
dc.subjectTraining trialsen_US
dc.subjectAlgorithmsen_US
dc.subjectFeature extractionen_US
dc.subjectLearning systemsen_US
dc.subjectSignal processingen_US
dc.subjectElectroencephalographyen_US
dc.titleExtraction of sparse spatial filters using Oscillating Searchen_US
dc.typeConference Paperen_US
dc.departmentDepartment of Electrical and Electronics Engineeringen_US
dc.identifier.doi10.1109/MLSP.2012.6349752en_US
dc.publisherIEEEen_US
dc.contributor.bilkentauthorÇetin, A. Enis


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record