4
$\begingroup$

I am working on a project that uses EEG signals of the brain to identify emotional states. While surveying the literature, I came across several references where "derived features of bispectrum" are used as features for the purpose of classifying motor and emotional signals.

Although I have a list of these "derived features" at my disposal, I would like to have some intuition as to which of these features I should use and why consider a bispectrum in the first place. I would highly appreciate it if the physical significance of the bispectrum could be explained in relation to EEG signals.

What is EEG bispectrum?

Improve this question
$\endgroup$
6
  • $\begingroup$ Welcome to CogSci.SE and thanks for your interesting question! Could you please add a reference or article link describing these derived features of the bispectrum? $\endgroup$
    – AliceD ♦
    Commented Jan 31, 2015 at 6:35
  • 2
    $\begingroup$ Sure!! Introduction to bispectrum for EEG analysis, Using HOS from EEG signals for developing BCI @ChrisStronks $\endgroup$
    – Nitin Kumar
    Commented Jan 31, 2015 at 9:28
  • $\begingroup$ Thanks. And what exactly is the list, as you ask which of them you should use? $\endgroup$
    – AliceD ♦
    Commented Jan 31, 2015 at 11:05
  • $\begingroup$ The list is not central to the question, as there are methods of determining a favorable set of features for classification purposes. What I am not getting is, how the bispectrum of EEG would prove to be a better discriminant than other measures such as features of PSD of EEG. @ChrisStronks $\endgroup$
    – Nitin Kumar
    Commented Jan 31, 2015 at 11:18
  • $\begingroup$ I don't think it is a matter of better or worse, but to a matter of adding information. FFT results in frequency and phase data. Often the phase (50% of the FFT data in fact) is thrown away without bothering. $\endgroup$
    – AliceD ♦
    Commented Jan 31, 2015 at 14:00

0

Reset to default