[Eeglablist] Queries related to Feature Extraction from EEG data

[AKSHANSH GUPTA]

Have u tried some feature selection methos?


On Sat, May 17, 2014 at 6:10 PM, Laxmi Shaw <laxmi.shaw22 at gmail.com> wrote:

>
> Dear EEGLAB List,
> I have extracted many features from EEG data. Features like mean,
> variance, kurtosis, skewness,
> Hjorth parameters etc. I am working on 64 channel data containing 64000
> samples each. So per subject I have a 64000*64 data-set to deal with and I
> have considered 34 subjects.
>
> After standard pre-processing,I have adopted constant segmentation for
> each channel. The segment width is 1 second(256 samples) with 50 percent
> overlapping. I have calculated statistical parameters as mentioned above
> for these small segments. In this way I have tried to reduce the data
> dimensionality to some extent. But I am unable to form a feature vector
> that i can input to some standard classifier or clustering tool.
>
> Please help with this problem. Please tell me whether my procedure is fine
> or I need to make some amendments. Above all please, help me form a feature
> vector that i can input into a classifier.
> Hoping for a positive response.
>
> -Regards
> PhD Scholar
> IIT Kharagpur,
> West Bengal
> Ph no-08388837821
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 

*Thanks & Regards...Akshansh Gupta*

*(Pursuing Ph.d ,  J.N.U,New Delhi-67)Contact no:  +91 9013927309,
+91 11-26741787,9899435903*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20140519/8e908f9a/attachment.html>