EEG clasifier question regarding label selection

✉️ Mailing List Archive (read-only)
1

External Email - Use Caution

Hello All,

I am new to machine learning and python mne, but my interest is situated
around developing Supervised learning model using EEG data.

I have a question about the aspect of choosing a label.

Do i have to choose one feature as my label
or
Do i have to enter manually digital representation for my labels?

For example

I have collected EEG data during two condition experiment (decision making
under low risk and decision making under high-risk condition)

My labels here are high and low risk
How do I represent this during my model development

Also, can someone point me to how to some feature selection examples,
having done the feature extraction?

looking forward to your reply

A. Ighoyota ben
Junior Researcher HCI (PhD in-view)
Tallinn University, Estonia
School of digital Technologies.
mobile:+372582 <+372%205832%206393>78794
skype: ighoyota-ben

<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon>
Virus-free.
www.avast.com
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20190220/b25a1b18/attachment.html

2

External Email - Use Caution

Hi Ighoyota ben
Providing a class label for Machine learning is fully depends on you. For
binary classification generally, labels are 0 and 1.
For your task, classification of low-risk and high-risk class labels
assignment based on your work (what you want to predict). For example, if
your work is finding high-risk EEG, then the high-risk class is a positive
class. Assing '1' for the class label for high-risk related features.

F1

Class

Low-risk (take as 0)

high-risk (take as 1)
for more clarification about positive class and negative class
https://developers.google.com/machine-learning/crash-course/classification/true-false-positive-negative