-1
$\begingroup$

i have 16 channel 2 class motor imaginary right hand ,right leg data.IS it possible to classify this through multiple linear regression model based on feature extraction?can i classify 3 class data like LF,RH,RF

$\endgroup$

closed as unclear what you're asking by Marcus Müller, lennon310, Stanley Pawlukiewicz, MBaz, A_A Sep 3 '18 at 10:02

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ sorry, this is still unclear and too broad. You can classify anything, sometimes even meaningfully, but it's totally unclear what features you want to extract to from what. We're not in your head! You have to imagine (or not, because this is the actual case) that you're explaining your question to someone who's never heard of what you're doing.We're experts, but not for your work, but for signal processing. Please learn to tell common knowledge from things only people can answer that are you or your advisor. Considering the quality of your previous questions, have you talked to your advisor? $\endgroup$ – Marcus Müller Sep 1 '18 at 9:05
  • $\begingroup$ Yes i have talked to my adviser.actually,i have extracted some features like median,mode,standard deviation from the EEG signal.i want to train the multiple linear regression model for classify or signal recognition .Is it possible or not? $\endgroup$ – galib29 Sep 1 '18 at 14:47
  • $\begingroup$ @galib29 Yes it is, in theory at least. What exactly are you interested in? Can you make the question a bit more specific? $\endgroup$ – A_A Sep 3 '18 at 10:02
0
$\begingroup$

multiple linear regression produces a linear model of a response as a weighted sum of predictor variables.

I believe that you are asking an equivalent question, that is, if a linear classifier can classify.

The answer would be yes, qualified to mean as well as a linear classifier can.

The linear classifier has limitations such as discussed in:

https://en.m.wikipedia.org/wiki/Perceptron

where one example is known as the xor problem.

I would think that given the data you have, you could answer your own question.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.