Feature Extraction and Classification of Biosignals - Emotion Valence Detection from EEG Signals

Tomé, A. M.; Hidalgo-Muñoz, A.R.; López, M.M.; Teixeira, Ana; Santos, I.M.; Pereira, A.T.; Vázquez-Marrufo, M.; Lang, E.W.

http://hdl.handle.net/10400.26/47375

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
Feature Extraction and Classification of Biosignals.pdf		191.92 KB	Adobe PDF	Download

Send Feedback

Authors

Abstract(s)

In thisworkavalencerecognitionsystembasedonelectroencephalogramsispresented.Theperformanceof the systemisevaluatedfortwosettings:singlesubjects(intra-subject)andbetweensubjects(inter-subject). The featureextractionisbasedonmeasuresofrelativeenergiescomputedinshorttimeintervalsandcertain frequencybands.Thefeatureextractionisperformedeitheronsignalsaveragedoveranensembleoftrialsor on single-trialresponsesignals.Thesubsequentclassificationstageisbasedonanensembleclassifier,i.e.a random forestoftreeclassifiers.Theclassificationisperformedconsideringtheensembleaverageresponsesof all subjects(inter-subject)orconsideringthesingle-trialresponsesofsinglesubjects(intra-subject).Applying a properimportancemeasureoftheclassifier,featureeliminationhasbeenusedtoidentifythemostrelevant features of the decision making.