emotopeaceproject.blogspot.com ameblo.jp 水の伝道師・江本勝が送るメッセージ2011年3...
Chris Pearson, visiting fellow at The Open University based at Rutherford Appleton Laboratory, gives...
The Montgomery Endowment presents John Burns "Five Years in Iraq: Which Way Home?" October 21, 2008...
Honorary Fellow-in-Residence Nan Aron, President of Alliance for Justice, examines the importance of...
Bat Boy infiltrated the beginning of EC10 lecture on November 30, 2009. www.batboylives.com
citeseer |
(0) (0 Votes)
|
Views: (1054) Date: (08-04-09) Pages: () |
Abstract: We consider Magnetoencephalograp hic (MEG) data in a signal detection framework. Our data set consists of responses evoked by the voiced syllables /b? / and /d? / and the corresponding voiceless syllables /p?/ and /t?/. The data yield well to principal component analysis (PCA), with a reasonable subspace in the order of three components out of 37 channels. To discriminate between responses to the voiced and voiceless versions of a consonant we form a feature vector by either matched ?ltering or wavelet packet decomposition and use a mixture-of-experts model to classify the stimuli. Both choices of a feature vector lead to a signi?cant detection accuracy. Furthermore, we show how to estimate the onset time of a stimulus from a continuous data stream. 1